# K-Dense Web Use Case ROI Comparison

A side-by-side cost and time analysis of every public K-Dense Web example use case versus what it would cost a human professional to produce the same deliverable.

**Total use cases analyzed:** 67
**Source of truth:** [`src/data/use-cases.ts`](../src/data/use-cases.ts)

---

## Executive Summary

- **Total use cases:** 67
- **Aggregate human cost (consultant rates):** $3,087,800
- **Aggregate K-Dense Web cost (Plus plan):** $5,670
- **Aggregate cost savings:** $3,082,130 (**99.8% lower**)
- **Aggregate human wall-clock time:** 14,568 hours = **364 person-weeks** at 40 hr/week
- **Aggregate K-Dense Web wall-clock:** 166.8 hours total (10,005 minutes)
- **Cost multiplier (human / K-Dense), aggregate:** **545×**
- **Cost multiplier, median per use case:** 560×
- **Cost multiplier, mean per use case:** 596×
- **Time compression, median per use case:** 107×
- **Time compression, mean per use case:** 131×
- **Time compression, aggregate:** 87×

### Headline examples (highest absolute savings)

| Deliverable | Human cost | K-Dense cost | Multiplier | Time saved |
|-------------|-----------:|-------------:|-----------:|-----------:|
| [CoreWeave VC Due Diligence](#49-coreweave-vc-due-diligence--venture-capital) (63-page DD, 12 figures) | $136,800 | $150 | **912×** | 11.4 weeks → 5 hr |
| [Structure Therapeutics Investment DD](#48-structure-therapeutics-investment-due-diligence--biotech-investment) (52-page biotech DD, 16 figures) | $121,200 | $200 | **606×** | 10.1 weeks → 6 hr |
| [Carbon Removal Investment Pipeline](#37-carbon-removal-investment-pipeline--climate-tech) (DAC/OAE pipeline + memo) | $107,100 | $160 | **670×** | 11.9 weeks → 2 hr |
| [Ramp VC Due Diligence](#50-ramp-technologies-vc-due-diligence--venture-capital) (44-page DD) | $98,400 | $150 | **656×** | 8.2 weeks → 4.5 hr |
| [Viking Therapeutics Investment Analysis](#51-viking-therapeutics-vk2735-investment-analysis--biotech-investment) (catalyst report) | $78,000 | $100 | **780×** | 6.5 weeks → 2.3 hr |
| [CT-388 Competitive Intelligence](#64-ct-388-obesity-drug-competitive-intelligence--biotech-investment) (44-page CI) | $76,800 | $150 | **512×** | 6.4 weeks → 5 hr |
| [Restaurant Industry Investment Memo](#53-restaurant-industry-investment-analysis--equity-research) (32-page pair-trade memo) | $64,900 | $100 | **649×** | 5.9 weeks → 3.3 hr |
| [Sector Rotation Macro Strategy](#55-sector-rotation-in-fed-rate-cut-cycles--macro-strategy) (31-page memo, 15 figures) | $65,000 | $100 | **650×** | 6.5 weeks → 5 hr |
| [ARPA-H Policy Report](#65-arpa-h-health-innovation-policy-report--public-policy) (42-page Brookings-grade) | $61,600 | $185 | **333×** | 7.7 weeks → 4 hr |
| [Cybersecurity M&A Screening](#54-cybersecurity-ma-target-screening--ma-analysis) (14-page memo, 9 figures) | $59,800 | $50 | **1,196×** | 4.6 weeks → 75 min |

### Highest cost multipliers (best $ → $ leverage)

| Deliverable | Multiplier |
|-------------|-----------:|
| [Cybersecurity M&A Target Screening](#54-cybersecurity-ma-target-screening--ma-analysis) | **1,196×** |
| [HER2 ADC Competitive Intelligence](#57-her2-adc-competitive-intelligence--biotech-investment) | **1,176×** |
| [Caribbean CBI Tax Strategy](#21-caribbean-cbi-tax-strategy--financelegal) | **1,139×** |
| [ADC Oncology Pair Trade](#60-adc-oncology-pair-trade-analysis--biotech-investment) | **1,080×** |
| [GLP-1 Receptor Agonists Safety Analysis](#58-glp-1-receptor-agonists-safety-analysis--biotech-investment) | **1,008×** |
| [Chegg Short Thesis](#52-chegg-chgg-short-thesis-evaluation--equity-research) | **946×** |
| [CoreWeave VC Due Diligence](#49-coreweave-vc-due-diligence--venture-capital) | **912×** |

### Category subtotals

| Category | # Cases | Human cost | K-Dense cost | Savings | Multiplier |
|----------|--------:|-----------:|-------------:|--------:|-----------:|
| Biotech Investment | 8 | $528,000 | $750 | $527,250 | 704× |
| Cancer Biology | 7 | $251,300 | $493 | $250,807 | 510× |
| Venture Capital | 2 | $235,200 | $300 | $234,900 | 784× |
| Climate Tech | 2 | $162,000 | $280 | $161,720 | 579× |
| Environment | 3 | $129,600 | $336 | $129,264 | 386× |
| Genomics | 3 | $119,000 | $252 | $118,748 | 472× |
| Equity Research | 2 | $112,200 | $150 | $112,050 | 748× |
| Engineering | 2 | $105,600 | $195 | $105,405 | 542× |
| Macro Strategy | 2 | $102,000 | $150 | $101,850 | 680× |
| AI/Technical | 2 | $89,600 | $238 | $89,362 | 376× |
| Agriculture | 2 | $72,800 | $130 | $72,670 | 560× |
| Social Science | 2 | $65,800 | $140 | $65,660 | 470× |
| Public Policy | 1 | $61,600 | $185 | $61,415 | 333× |
| M&A Analysis | 1 | $59,800 | $50 | $59,750 | 1196× |
| Energy | 1 | $56,000 | $110 | $55,890 | 509× |
| Transportation | 1 | $54,400 | $105 | $54,295 | 518× |
| Economics | 1 | $42,400 | $50 | $42,350 | 848× |
| Quantum Chemistry | 1 | $42,300 | $50 | $42,250 | 846× |
| Maritime History | 1 | $42,000 | $95 | $41,905 | 442× |
| Finance/Legal | 1 | $41,000 | $36 | $40,964 | 1139× |
| Drug Discovery | 1 | $40,600 | $100 | $40,500 | 406× |
| Astronomy | 1 | $40,000 | $100 | $39,900 | 400× |
| Medical Devices | 1 | $39,900 | $99 | $39,801 | 403× |
| Scientometrics | 1 | $39,200 | $100 | $39,100 | 392× |
| Metabolomics | 1 | $37,800 | $79 | $37,721 | 478× |
| Biology/Safety | 1 | $37,800 | $50 | $37,750 | 756× |
| Neuroscience | 1 | $37,100 | $86 | $37,014 | 431× |
| Biosignals | 1 | $36,400 | $50 | $36,350 | 728× |
| Epigenetics | 1 | $35,700 | $50 | $35,650 | 714× |
| Robotics | 1 | $35,200 | $100 | $35,100 | 352× |
| Tourism/Economics | 1 | $35,000 | $150 | $34,850 | 233× |
| Nutrition | 1 | $30,100 | $100 | $30,000 | 301× |
| Biomedical Materials | 1 | $29,400 | $50 | $29,350 | 588× |
| Synthetic Biology | 1 | $28,700 | $50 | $28,650 | 574× |
| Toxicology | 1 | $28,700 | $50 | $28,650 | 574× |
| Market Research | 1 | $28,000 | $50 | $27,950 | 560× |
| Travel | 1 | $27,500 | $85 | $27,415 | 324× |
| Pharma/IP | 1 | $27,300 | $50 | $27,250 | 546× |
| Mass Spectrometry | 1 | $27,300 | $40 | $27,260 | 682× |
| Proteomics | 1 | $26,600 | $40 | $26,560 | 665× |
| Nanomedicine | 1 | $25,200 | $46 | $25,154 | 548× |
| Demographics | 1 | $21,700 | $50 | $21,650 | 434× |

---

## Methodology

### What this comparison is

For each public K-Dense Web example, we estimate **what it would have cost and how long it would have taken a single qualified human professional** (or small team) to produce the same final artifact (paper, report, slide deck, manuscript, decision memo, dashboard) and the same intermediate work (data acquisition, ML pipelines, figures, references), and we compare that to the actual K-Dense Web wall-clock and credit cost.

### What is counted in human cost

We sum hours across the line items the human would actually spend, then multiply by a domain-appropriate hourly rate:

1. Scoping and literature review
2. Data acquisition, cleaning, harmonization
3. Analysis, modeling, ML pipelines, statistics
4. Figure generation (typically 1–3 hrs per polished figure × N figures)
5. Writing (typically 1.5–3 hrs per final page × N pages, including LaTeX, BibTeX, formatting)
6. Peer review / revision pass (~10–20 % of writing time)

### What is **not** counted in human cost

We deliberately exclude:

- Data licensing fees (e.g., cBioPortal Pro, FAERS commercial licenses, premium news/IP databases)
- Software licenses (MATLAB, ChemDraw, COMSOL, Bloomberg, Capital IQ)
- Project management overhead and PI/principal review time
- Calendar overhead (queueing, meetings, dependencies)

These would all _widen_ the gap in K-Dense Web's favor, so this is a conservative comparison.

### Human hourly rates (mid-market US 2025 consulting / contract)

| Domain                                                            | $/hr |
|-------------------------------------------------------------------|------|
| Computational biology / bioinformatics / drug discovery / genomics | 175  |
| ML / data science / quant analysis                                 | 200  |
| VC due diligence (associate + partner blend)                       | 300  |
| Equity research / sell-side analyst                                | 275  |
| Macro strategy / economist                                         | 250  |
| M&A advisory                                                       | 325  |
| Public-policy analyst (think-tank / GAO-equivalent)                | 200  |
| Engineering analysis (structural, thermal, energy)                 | 200  |
| Climate tech / techno-economic analysis                            | 225  |
| Market research                                                    | 175  |
| Travel / lifestyle planning concierge                              | 125  |
| Social science / demographics                                      | 175  |
| Maritime / historical research                                     | 150  |
| AI / technical writing                                             | 200  |
| Nutrition / dietetics                                              | 150  |
| Tourism economics / strategy                                       | 175  |
| Toxicology / regulatory science                                    | 200  |
| Astronomy / observational science                                  | 200  |
| Quantum chemistry / computational physics                          | 225  |

These are blended individual-contributor rates, not BCG / McKinsey or specialty-firm fixed-fee deliverables (which would push the gap **3–10× wider** in K-Dense Web's favor).

### K-Dense Web cost basis

We use the **Plus plan effective price**: `$199 / 300 credits = $0.66 per credit`. This is what a real Plus subscriber (the recommended plan) actually pays per credit. Pay-as-you-go (Personal plan) is roughly 50 % more, and Team plan is slightly cheaper per credit.

Per-task credit estimates by complexity (used to set each row):

| Tier   | Description                                                                  | Credits   | $ at Plus rate |
|--------|------------------------------------------------------------------------------|-----------|----------------|
| Light  | Single chart / short memo / 1–2 page brief                                   | 10–25     | $7–17          |
| Medium | Multi-step analysis, ~10–20 page report, ≤6 figures                          | 40–80     | $27–53         |
| Heavy  | Full ML pipeline + 30–60 page paper, 8–14 figures                            | 100–200   | $66–132        |
| Mega   | 40+ page policy / DD report, 12+ figures, large ref list, multi-stage review | 200–300   | $132–198       |

K-Dense Web wall-clock is taken from the session manifest where stated (e.g. `~45 min execution time`), or estimated from output volume (typical range: 30 min – 6 hrs).

### Wall-clock vs calendar time

Human-hour estimates assume a **single specialist working full-time without context switching**. Calendar time in practice would be **2–5× longer** (waiting on data access, peer review, vendor turnaround, scheduling). K-Dense Web wall-clock is the actual end-to-end runtime.

### Caveats

- Estimates are order-of-magnitude (±30 %), not bid-grade.
- We use a single rate per row; real teams would mix junior/senior labor at different rates.
- For finance / investment use cases, real Wall Street DD reports often have list prices of **$50K–$500K** from boutiques. We use IC-rate × hours instead, which is a much more conservative number.
- For policy / NGO reports, real comparable deliverables (Brookings, RAND, GAO) are typically **$80K–$300K**. We again use the IC-rate floor.

---

## Master comparison table (67 use cases, ordered as in the gallery)

| # | Use Case | Category | Human time | Human cost | K-Dense time | K-Dense cost | Cost saved | Multiplier | Links |
|---|----------|----------|-----------:|-----------:|-------------:|-------------:|-----------:|-----------:|-------|
| 1 | [Natural Products Chemical Space](/blog/antimicrobial-drug-discovery-natural-products) | Drug Discovery | 5.8 weeks | $40,600 | 45 min | $100 | $40,500 | 406× | [Session](https://app.k-dense.ai/share/session_20251212_135154_dad1cf0d70ee) · [PDF](/examples/session_20251212_135154_dad1cf0d70ee/writing_outputs/final/NP_Antimicrobial_Screening_Manuscript.pdf) · [Blog](/blog/antimicrobial-drug-discovery-natural-products) |
| 2 | [Macroeconomic Recession Prediction](/examples/session_20251228_135128_cd3e46eedd35/writing_outputs/final/recession_early_warning_paper.pdf) | Economics | 5.3 weeks | $42,400 | 90 min | $50 | $42,350 | 848× | [Session](https://app.k-dense.ai/share/session_20251228_135128_cd3e46eedd35) · [PDF](/examples/session_20251228_135128_cd3e46eedd35/writing_outputs/final/recession_early_warning_paper.pdf) |
| 3 | [Longevity Gene Analysis](/examples/session_20251215_165944_a1f9981a1a45/writing_outputs/final/longevity_translation_scorecard.pdf) | Genomics | 4.7 weeks | $32,900 | 75 min | $50 | $32,850 | 658× | [Session](https://app.k-dense.ai/share/session_20251215_165944_a1f9981a1a45) · [PDF](/examples/session_20251215_165944_a1f9981a1a45/writing_outputs/final/longevity_translation_scorecard.pdf) |
| 4 | [JWST Exoplanet Target Prioritization](/examples/session_20251214_135820_a55c35bcb856/writing_outputs/final/JWST_Biosignature_Priority_Framework.pdf) | Astronomy | 5.0 weeks | $40,000 | 2.5 hr | $100 | $39,900 | 400× | [Session](https://app.k-dense.ai/share/session_20251214_135820_a55c35bcb856) · [PDF](/examples/session_20251214_135820_a55c35bcb856/writing_outputs/final/JWST_Biosignature_Priority_Framework.pdf) |
| 5 | [Quantum Chemistry VQE Benchmarking](/examples/session_20251212_162938_43788a48ae9a/writing_outputs/final/vqe_benchmarking_paper.pdf) | Quantum Chemistry | 4.7 weeks | $42,300 | 2.0 hr | $50 | $42,250 | 846× | [Session](https://app.k-dense.ai/share/session_20251212_162938_43788a48ae9a) · [PDF](/examples/session_20251212_162938_43788a48ae9a/writing_outputs/final/vqe_benchmarking_paper.pdf) |
| 6 | [Doxorubicin Mechanism of Action](https://app.k-dense.ai/share/session_20251217_123457_77eda5efa279) | Cancer Biology | 5.4 weeks | $37,800 | 2.0 hr | $50 | $37,750 | 756× | [Session](https://app.k-dense.ai/share/session_20251217_123457_77eda5efa279) |
| 7 | [ECG Stress Detection Pipeline](/examples/session_20251212_210544_328dfde1326e/writing_outputs/final/hrv_stress_detection_manuscript.pdf) | Biosignals | 5.2 weeks | $36,400 | 2.0 hr | $50 | $36,350 | 728× | [Session](https://app.k-dense.ai/share/session_20251212_210544_328dfde1326e) · [PDF](/examples/session_20251212_210544_328dfde1326e/writing_outputs/final/hrv_stress_detection_manuscript.pdf) |
| 8 | [Lycopene Biosynthesis Optimization](/examples/session_20251215_170752_24b906eb7e01/writing_outputs/final/manuscript.pdf) | Synthetic Biology | 4.1 weeks | $28,700 | 90 min | $50 | $28,650 | 574× | [Session](https://app.k-dense.ai/share/session_20251215_170752_24b906eb7e01) · [PDF](/examples/session_20251215_170752_24b906eb7e01/writing_outputs/final/manuscript.pdf) |
| 9 | [Epigenetic Clock Drug Discovery](/examples/session_20251215_170449_b3d275ad14ce/writing_outputs/final/epigenetic_aging_mechanisms.pdf) | Epigenetics | 5.1 weeks | $35,700 | 2.0 hr | $50 | $35,650 | 714× | [Session](https://app.k-dense.ai/share/session_20251215_170449_b3d275ad14ce) · [PDF](/examples/session_20251215_170449_b3d275ad14ce/writing_outputs/final/epigenetic_aging_mechanisms.pdf) |
| 10 | [GLP-1 Drug Patent Landscape](/examples/session_20251214_114340_87efd8b52064/writing_outputs/final/patent_vulnerability_report.pdf) | Pharma/IP | 3.9 weeks | $27,300 | 75 min | $50 | $27,250 | 546× | [Session](https://app.k-dense.ai/share/session_20251214_114340_87efd8b52064) · [PDF](/examples/session_20251214_114340_87efd8b52064/writing_outputs/final/patent_vulnerability_report.pdf) |
| 11 | [Biodegradable Polymer Safety Analysis](/examples/session_20251212_210923_d71aa9ce43f6/writing_outputs/final/manuscript.pdf) | Biomedical Materials | 4.2 weeks | $29,400 | 90 min | $50 | $29,350 | 588× | [Session](https://app.k-dense.ai/share/session_20251212_210923_d71aa9ce43f6) · [PDF](/examples/session_20251212_210923_d71aa9ce43f6/writing_outputs/final/manuscript.pdf) |
| 12 | [ICD Adverse Events Analysis](/examples/session_20251212_210203_5e688fa2ddca/writing_outputs/final/ICD_Adverse_Events_Technical_Report.pdf) | Medical Devices | 5.7 weeks | $39,900 | 3.0 hr | $99 | $39,800 | 403× | [Session](https://app.k-dense.ai/share/session_20251212_210203_5e688fa2ddca) · [PDF](/examples/session_20251212_210203_5e688fa2ddca/writing_outputs/final/ICD_Adverse_Events_Technical_Report.pdf) |
| 13 | [Nanoparticle Protein Corona Prediction](https://app.k-dense.ai/share/session_20251214_140128_b84388b8e99e) | Nanomedicine | 3.6 weeks | $25,200 | 80 min | $46 | $25,150 | 548× | [Session](https://app.k-dense.ai/share/session_20251214_140128_b84388b8e99e) |
| 14 | [MS Fragmentation Rule Mining](/examples/session_20251215_165434_bcb1a47df161/writing_outputs/final/fragmentation_rules_paper.pdf) | Mass Spectrometry | 3.9 weeks | $27,300 | 90 min | $40 | $27,250 | 683× | [Session](https://app.k-dense.ai/share/session_20251215_165434_bcb1a47df161) · [PDF](/examples/session_20251215_165434_bcb1a47df161/writing_outputs/final/fragmentation_rules_paper.pdf) |
| 15 | [Wearable Seizure Prediction](/examples/session_20251214_140801_5fd244475cff/writing_outputs/final/manuscript.pdf) | Neuroscience | 5.3 weeks | $37,100 | 2.5 hr | $86 | $37,000 | 431× | [Session](https://app.k-dense.ai/share/session_20251214_140801_5fd244475cff) · [PDF](/examples/session_20251214_140801_5fd244475cff/writing_outputs/final/manuscript.pdf) |
| 16 | [Air Quality Health Impact Analysis](https://app.k-dense.ai/share/session_20251212_124112_f9f3e9939b7d) | Environment | 3.6 weeks | $28,800 | 90 min | $36 | $28,750 | 800× | [Session](https://app.k-dense.ai/share/session_20251212_124112_f9f3e9939b7d) |
| 17 | [LC-MS Ion Suppression Prediction](/examples/session_20251215_164703_dc8bf4ef0471/writing_outputs/final/ion_suppression_paper.pdf) | Metabolomics | 5.4 weeks | $37,800 | 3.0 hr | $79 | $37,700 | 478× | [Session](https://app.k-dense.ai/share/session_20251215_164703_dc8bf4ef0471) · [PDF](/examples/session_20251215_164703_dc8bf4ef0471/writing_outputs/final/ion_suppression_paper.pdf) |
| 18 | [Agent Skills vs MCP Architecture](/examples/session_20251231_185247_6dce8fea6faa/writing_outputs/final/agent_skills_vs_mcp_report.pdf) | AI/Technical | 3.4 weeks | $27,200 | 105 min | $43 | $27,150 | 633× | [Session](https://app.k-dense.ai/share/session_20251231_185247_6dce8fea6faa) · [PDF](/examples/session_20251231_185247_6dce8fea6faa/writing_outputs/final/agent_skills_vs_mcp_report.pdf) |
| 19 | [TMT Proteomics Spike-in Validation](/examples/session_20251212_165231_ad0681ba0d19/writing_outputs/final/Grant_Report.pdf) | Proteomics | 3.8 weeks | $26,600 | 75 min | $40 | $26,550 | 665× | [Session](https://app.k-dense.ai/share/session_20251212_165231_ad0681ba0d19) · [PDF](/examples/session_20251212_165231_ad0681ba0d19/writing_outputs/final/Grant_Report.pdf) |
| 20 | [T2-Low Asthma Endotype Discovery](/examples/session_20251216_063845_10f14a0ac5fe/writing_outputs/final/Asthma_Endotyping_Presentation.pdf) | Genomics | 4.9 weeks | $34,300 | 3.0 hr | $102 | $34,200 | 336× | [Session](https://app.k-dense.ai/share/session_20251216_063845_10f14a0ac5fe) · [PDF](/examples/session_20251216_063845_10f14a0ac5fe/writing_outputs/final/Asthma_Endotyping_Presentation.pdf) |
| 21 | [Caribbean CBI Tax Strategy](/examples/session_20251217_072723_efc7231a0106/writing_outputs/final/executive_decision_report.pdf) | Finance/Legal | 4.1 weeks | $41,000 | 90 min | $36 | $40,950 | 1.1K× | [Session](https://app.k-dense.ai/share/session_20251217_072723_efc7231a0106) · [PDF](/examples/session_20251217_072723_efc7231a0106/writing_outputs/final/executive_decision_report.pdf) |
| 22 | [HNSCC Treatment Response Biomarkers](https://app.k-dense.ai/share/session_20251217_122112_5804b26745cf) | Cancer Biology | 4.0 weeks | $28,000 | 105 min | $43 | $27,950 | 651× | [Session](https://app.k-dense.ai/share/session_20251217_122112_5804b26745cf) |
| 23 | [Tox21 Liver Toxicity Prediction](/examples/session_20251217_122902_9ad327069ee6/writing_outputs/final/predictive_toxicology_paper.pdf) | Toxicology | 4.1 weeks | $28,700 | 75 min | $50 | $28,650 | 574× | [Session](https://app.k-dense.ai/share/session_20251217_122902_9ad327069ee6) · [PDF](/examples/session_20251217_122902_9ad327069ee6/writing_outputs/final/predictive_toxicology_paper.pdf) |
| 24 | [US Vineyard Location Analysis](/examples/session_20251218_143137_2f4bd2ba43ea/writing_outputs/final/vineyard_location_report.pdf) | Agriculture | 3.7 weeks | $25,900 | 90 min | $50 | $25,850 | 518× | [Session](https://app.k-dense.ai/share/session_20251218_143137_2f4bd2ba43ea) · [PDF](/examples/session_20251218_143137_2f4bd2ba43ea/writing_outputs/final/vineyard_location_report.pdf) |
| 25 | [Aravali Mountains Impact Prediction](/examples/session_20251219_144034_6ca0fee97d6e/writing_outputs/final/aravali_crisis_research_paper.pdf) | Environment | 6.2 weeks | $49,600 | 6.0 hr | $150 | $49,450 | 331× | [Session](https://app.k-dense.ai/share/session_20251219_144034_6ca0fee97d6e) · [PDF](/examples/session_20251219_144034_6ca0fee97d6e/writing_outputs/final/aravali_crisis_research_paper.pdf) |
| 26 | [Indian Vegetarian Diet Optimization](/examples/session_20251221_093619_3390c5c3bed6/writing_outputs/final/india_vegetarian_diet_plan.pdf) | Nutrition | 4.3 weeks | $30,100 | 2.0 hr | $100 | $30,000 | 301× | [Session](https://app.k-dense.ai/share/session_20251221_093619_3390c5c3bed6) · [PDF](/examples/session_20251221_093619_3390c5c3bed6/writing_outputs/final/india_vegetarian_diet_plan.pdf) |
| 27 | [East vs West Cultural Analysis](/examples/session_20251223_092351_e1527e79a0d2/writing_outputs/final/presentation.pdf) | Social Science | 2.3 weeks | $16,100 | 45 min | $50 | $16,050 | 322× | [Session](https://app.k-dense.ai/share/session_20251223_092351_e1527e79a0d2) · [PDF](/examples/session_20251223_092351_e1527e79a0d2/writing_outputs/final/presentation.pdf) |
| 28 | [Renal Cancer Drug Response Stratification](https://app.k-dense.ai/share/session_20251224_080443_4cf56e41bc8e) | Cancer Biology | 7.2 weeks | $50,400 | 4.0 hr | $100 | $50,300 | 504× | [Session](https://app.k-dense.ai/share/session_20251224_080443_4cf56e41bc8e) |
| 29 | [Indian State Emigration Patterns](/examples/session_20251224_091022_74d424b00564/writing_outputs/final/Indian_Expat_Analysis_Presentation.pdf) | Demographics | 3.1 weeks | $21,700 | 75 min | $50 | $21,650 | 434× | [Session](https://app.k-dense.ai/share/session_20251224_091022_74d424b00564) · [PDF](/examples/session_20251224_091022_74d424b00564/writing_outputs/final/Indian_Expat_Analysis_Presentation.pdf) |
| 30 | [50m Concrete Dome Structural Analysis](/examples/session_20251224_102759_927c1cef200e/writing_outputs/final/50m_dome_technical_report.pdf) | Engineering | 5.8 weeks | $46,400 | 3.0 hr | $100 | $46,300 | 464× | [Session](https://app.k-dense.ai/share/session_20251224_102759_927c1cef200e) · [PDF](/examples/session_20251224_102759_927c1cef200e/writing_outputs/final/50m_dome_technical_report.pdf) |
| 31 | [India Urban Air Pollution Tipping Points](/examples/session_20251224_200424_4a6cc33ed0fd/writing_outputs/final/india_pollution_research_paper.pdf) | Environment | 6.4 weeks | $51,200 | 5.0 hr | $150 | $51,050 | 341× | [Session](https://app.k-dense.ai/share/session_20251224_200424_4a6cc33ed0fd) · [PDF](/examples/session_20251224_200424_4a6cc33ed0fd/writing_outputs/final/india_pollution_research_paper.pdf) |
| 32 | [India vs China AI Research Analysis](/examples/session_20251226_095001_6ed1c15df75b/writing_outputs/final/manuscript.pdf) | Scientometrics | 4.9 weeks | $39,200 | 2.5 hr | $100 | $39,100 | 392× | [Session](https://app.k-dense.ai/share/session_20251226_095001_6ed1c15df75b) · [PDF](/examples/session_20251226_095001_6ed1c15df75b/writing_outputs/final/manuscript.pdf) |
| 33 | [India Sacred Tree Tourism Strategy](/examples/session_20251227_090118_524452720570/writing_outputs/final/indian_heritage_tree_tourism_strategic_analysis.pdf) | Tourism/Economics | 5.0 weeks | $35,000 | 6.0 hr | $150 | $34,850 | 233× | [Session](https://app.k-dense.ai/share/session_20251227_090118_524452720570) · [PDF](/examples/session_20251227_090118_524452720570/writing_outputs/final/indian_heritage_tree_tourism_strategic_analysis.pdf) |
| 34 | [India Crop Yield Prediction](/examples/session_20251229_081931_9f2d364070e9/writing_outputs/final/indian_agriculture_analysis.pdf) | Agriculture | 6.7 weeks | $46,900 | 3.0 hr | $80 | $46,800 | 597× | [Session](https://app.k-dense.ai/share/session_20251229_081931_9f2d364070e9) · [PDF](/examples/session_20251229_081931_9f2d364070e9/writing_outputs/final/indian_agriculture_analysis.pdf) |
| 35 | [Solar Green Hydrogen Thermal Analysis](/examples/session_20251230_095716_0b529465c6eb/writing_outputs/final/green_hydrogen_paper.pdf) | Energy | 7.0 weeks | $56,000 | 3.2 hr | $110 | $55,850 | 527× | [Session](https://app.k-dense.ai/share/session_20251230_095716_0b529465c6eb) · [PDF](/examples/session_20251230_095716_0b529465c6eb/writing_outputs/final/green_hydrogen_paper.pdf) |
| 36 | [India EV Adoption Econometrics](/examples/session_20251230_101850_14dd57dfe39b/writing_outputs/final/charging_ahead_ev_india.pdf) | Transportation | 6.8 weeks | $54,400 | 2.5 hr | $105 | $54,250 | 522× | [Session](https://app.k-dense.ai/share/session_20251230_101850_14dd57dfe39b) · [PDF](/examples/session_20251230_101850_14dd57dfe39b/writing_outputs/final/charging_ahead_ev_india.pdf) |
| 37 | [Carbon Removal Investment Pipeline](https://app.k-dense.ai/share/session_20251212_164706_1df6aaa34e0b) | Climate Tech | 11.9 weeks | $107,100 | 2.0 hr | $160 | $106,950 | 670× | [Session](https://app.k-dense.ai/share/session_20251212_164706_1df6aaa34e0b) |
| 38 | [Infrastructure Impact on Women's Empowerment](/examples/session_20251229_112053_41c35bf38d4f/writing_outputs/final/infrastructure_gender_research_v2.pdf) | Social Science | 7.1 weeks | $49,700 | 2.5 hr | $90 | $49,550 | 553× | [Session](https://app.k-dense.ai/share/session_20251229_112053_41c35bf38d4f) · [PDF](/examples/session_20251229_112053_41c35bf38d4f/writing_outputs/final/infrastructure_gender_research_v2.pdf) |
| 39 | [NGS Variant Callers in Clinical Practice](/examples/session_20251224_173534_9bb5ad16d19b/writing_outputs/final/NGS_Variant_Callers_Review.pdf) | Genomics | 7.4 weeks | $51,800 | 3.2 hr | $100 | $51,650 | 528× | [Session](https://app.k-dense.ai/share/session_20251224_173534_9bb5ad16d19b) · [PDF](/examples/session_20251224_173534_9bb5ad16d19b/writing_outputs/final/NGS_Variant_Callers_Review.pdf) |
| 40 | [New Zealand Great Walks Planning](/examples/session_20251224_112534_0ab5bab9278d/writing_outputs/final/nz_great_walks_trip_plan.pdf) | Travel | 5.5 weeks | $27,500 | 3.0 hr | $85 | $27,400 | 324× | [Session](https://app.k-dense.ai/share/session_20251224_112534_0ab5bab9278d) · [PDF](/examples/session_20251224_112534_0ab5bab9278d/writing_outputs/final/nz_great_walks_trip_plan.pdf) |
| 41 | [Flying Dutchman Historical Analysis](/examples/session_20251224_182332_0a76341055a3/writing_outputs/final/flying_dutchman_paper.pdf) | Maritime History | 7.0 weeks | $42,000 | 3.2 hr | $95 | $41,850 | 446× | [Session](https://app.k-dense.ai/share/session_20251224_182332_0a76341055a3) · [PDF](/examples/session_20251224_182332_0a76341055a3/writing_outputs/final/flying_dutchman_paper.pdf) |
| 42 | [Great Pyramid Construction Analysis](/examples/session_20251224_105449_4bd8fa278c83/writing_outputs/final/pyramid_engineering_analysis.pdf) | Engineering | 7.4 weeks | $59,200 | 3.2 hr | $95 | $59,100 | 623× | [Session](https://app.k-dense.ai/share/session_20251224_105449_4bd8fa278c83) · [PDF](/examples/session_20251224_105449_4bd8fa278c83/writing_outputs/final/pyramid_engineering_analysis.pdf) |
| 43 | [AI Predictions 2026](/examples/session_20251231_185649_7c104f74b40a/writing_outputs/final/AI_Predictions_2026.pdf) | AI/Technical | 7.8 weeks | $62,400 | 40 min | $195 | $62,150 | 321× | [Session](https://app.k-dense.ai/share/session_20251231_185649_7c104f74b40a) · [PDF](/examples/session_20251231_185649_7c104f74b40a/writing_outputs/final/AI_Predictions_2026.pdf) |
| 44 | [Deadliest Animal Venoms Safety Guide](/examples/session_20251224_181033_118bb9451f72/writing_outputs/final/v1_venom_guide.pdf) | Biology/Safety | 5.4 weeks | $37,800 | 90 min | $50 | $37,700 | 756× | [Session](https://app.k-dense.ai/share/session_20251224_181033_118bb9451f72) · [PDF](/examples/session_20251224_181033_118bb9451f72/writing_outputs/final/v1_venom_guide.pdf) |
| 45 | [PDAC KRAS Mutation Survival Analysis](/examples/session_20260116_102423_c4b0c6a77530/writing_outputs/final/manuscript.pdf) | Cancer Biology | 4.0 weeks | $28,000 | 90 min | $50 | $27,950 | 560× | [Session](https://app.k-dense.ai/share/session_20260116_102423_c4b0c6a77530) · [PDF](/examples/session_20260116_102423_c4b0c6a77530/writing_outputs/final/manuscript.pdf) |
| 46 | [Biologically Inspired Robot Actuators PhD Proposal](/blog/ai-research-direction-discovery-phd-proposal) | Robotics | 4.4 weeks | $35,200 | 45 min | $100 | $35,100 | 352× | [Session](https://app.k-dense.ai/share/session_20260119_113716_05b0fba1834d) · [PDF](/examples/session_20260119_113716_05b0fba1834d/writing_outputs/final/phd_research_proposal.pdf) · [Blog](/blog/ai-research-direction-discovery-phd-proposal) |
| 47 | [GBM Clinical Trial Landscape Analysis](/blog/gbm-clinical-trial-landscape-analysis) | Cancer Biology | 4.7 weeks | $32,900 | 100 min | $50 | $32,850 | 658× | [Session](https://app.k-dense.ai/share/session_20260122_114521_a550ad5331dd) · [PDF](/examples/session_20260122_114521_a550ad5331dd/writing_outputs/final/GBM_Clinical_Trial_Landscape_Analysis.pdf) · [Blog](/blog/gbm-clinical-trial-landscape-analysis) |
| 48 | [Structure Therapeutics Investment Due Diligence](/examples/session_20260127_141459_d78bbad6de57/writing_outputs/final/structure_therapeutics_due_diligence.pdf) | Biotech Investment | 10.1 weeks | $121,200 | 6.0 hr | $200 | $121,000 | 606× | [Session](https://app.k-dense.ai/share/session_20260127_141459_d78bbad6de57) · [PDF](/examples/session_20260127_141459_d78bbad6de57/writing_outputs/final/structure_therapeutics_due_diligence.pdf) |
| 49 | [CoreWeave VC Due Diligence](/examples/session_20260127_143018_1e043a3c21a2/writing_outputs/final/coreweave_due_diligence_report.pdf) | Venture Capital | 11.4 weeks | $136,800 | 5.0 hr | $150 | $136,650 | 912× | [Session](https://app.k-dense.ai/share/session_20260127_143018_1e043a3c21a2) · [PDF](/examples/session_20260127_143018_1e043a3c21a2/writing_outputs/final/coreweave_due_diligence_report.pdf) |
| 50 | [Ramp Technologies VC Due Diligence](/examples/session_20260127_143328_800725940a1f/writing_outputs/final/ramp_due_diligence.pdf) | Venture Capital | 8.2 weeks | $98,400 | 4.5 hr | $150 | $98,250 | 656× | [Session](https://app.k-dense.ai/share/session_20260127_143328_800725940a1f) · [PDF](/examples/session_20260127_143328_800725940a1f/writing_outputs/final/ramp_due_diligence.pdf) |
| 51 | [Viking Therapeutics (VK2735) Investment Analysis](/examples/session_20260203_085220_b017de51c5a6/writing_outputs/final/VKTX_Catalyst_Analysis_Report.pdf) | Biotech Investment | 6.5 weeks | $78,000 | 2.3 hr | $100 | $77,900 | 780× | [Session](https://app.k-dense.ai/share/session_20260203_085220_b017de51c5a6) · [PDF](/examples/session_20260203_085220_b017de51c5a6/writing_outputs/final/VKTX_Catalyst_Analysis_Report.pdf) |
| 52 | [Chegg (CHGG) Short Thesis Evaluation](/examples/session_20260203_085912_030e6e80a419/writing_outputs/final/CHGG_Short_Thesis.pdf) | Equity Research | 4.3 weeks | $47,300 | 60 min | $50 | $47,250 | 946× | [Session](https://app.k-dense.ai/share/session_20260203_085912_030e6e80a419) · [PDF](/examples/session_20260203_085912_030e6e80a419/writing_outputs/final/CHGG_Short_Thesis.pdf) |
| 53 | [Restaurant Industry Investment Analysis](/examples/session_20260203_085615_5f10d3d2357f/writing_outputs/final/investment_memo.pdf) | Equity Research | 5.9 weeks | $64,900 | 3.3 hr | $100 | $64,800 | 649× | [Session](https://app.k-dense.ai/share/session_20260203_085615_5f10d3d2357f) · [PDF](/examples/session_20260203_085615_5f10d3d2357f/writing_outputs/final/investment_memo.pdf) |
| 54 | [Cybersecurity M&A Target Screening](/examples/session_20260203_085757_d926b2c7d1fb/writing_outputs/final/cybersecurity_ma_screening_memo.pdf) | M&A Analysis | 4.6 weeks | $59,800 | 75 min | $50 | $59,750 | 1.2K× | [Session](https://app.k-dense.ai/share/session_20260203_085757_d926b2c7d1fb) · [PDF](/examples/session_20260203_085757_d926b2c7d1fb/writing_outputs/final/cybersecurity_ma_screening_memo.pdf) |
| 55 | [Sector Rotation in Fed Rate Cut Cycles](/examples/session_20260203_090059_a06c4b9b68c4/writing_outputs/final/macro_strategy_memo.pdf) | Macro Strategy | 6.5 weeks | $65,000 | 5.0 hr | $100 | $64,900 | 650× | [Session](https://app.k-dense.ai/share/session_20260203_090059_a06c4b9b68c4) · [PDF](/examples/session_20260203_090059_a06c4b9b68c4/writing_outputs/final/macro_strategy_memo.pdf) |
| 56 | [Resmetirom FDA Approval Probability](/examples/session_20260203_090952_ba7183dc7776/writing_outputs/final/fda_catalyst_brief_resmetirom.pdf) | Biotech Investment | 4.2 weeks | $50,400 | 3.3 hr | $100 | $50,300 | 504× | [Session](https://app.k-dense.ai/share/session_20260203_090952_ba7183dc7776) · [PDF](/examples/session_20260203_090952_ba7183dc7776/writing_outputs/final/fda_catalyst_brief_resmetirom.pdf) |
| 57 | [HER2 ADC Competitive Intelligence](/examples/session_20260203_091112_0888c802c3bf/writing_outputs/final/HER2_ADC_Competitive_Intelligence_Report.pdf) | Biotech Investment | 4.9 weeks | $58,800 | 100 min | $50 | $58,750 | 1.2K× | [Session](https://app.k-dense.ai/share/session_20260203_091112_0888c802c3bf) · [PDF](/examples/session_20260203_091112_0888c802c3bf/writing_outputs/final/HER2_ADC_Competitive_Intelligence_Report.pdf) |
| 58 | [GLP-1 Receptor Agonists Safety Analysis](/examples/session_20260203_091322_5981a70f834a/writing_outputs/final/GLP1_Safety_Signal_Report.pdf) | Biotech Investment | 4.2 weeks | $50,400 | 75 min | $50 | $50,350 | 1.0K× | [Session](https://app.k-dense.ai/share/session_20260203_091322_5981a70f834a) · [PDF](/examples/session_20260203_091322_5981a70f834a/writing_outputs/final/GLP1_Safety_Signal_Report.pdf) |
| 59 | [GLP-1 Adverse Events FAERS Analysis](/examples/session_20260203_092044_b80cb683d2ee/writing_outputs/final/GLP1_Safety_Analysis_Presentation.pdf) | Biotech Investment | 3.2 weeks | $38,400 | 70 min | $50 | $38,350 | 768× | [Session](https://app.k-dense.ai/share/session_20260203_092044_b80cb683d2ee) · [PDF](/examples/session_20260203_092044_b80cb683d2ee/writing_outputs/final/GLP1_Safety_Analysis_Presentation.pdf) |
| 60 | [ADC Oncology Pair Trade Analysis](/examples/session_20260203_092344_47a31388326a/writing_outputs/final/adc_oncology_pair_trade_report.pdf) | Biotech Investment | 4.5 weeks | $54,000 | 25 min | $50 | $53,950 | 1.1K× | [Session](https://app.k-dense.ai/share/session_20260203_092344_47a31388326a) · [PDF](/examples/session_20260203_092344_47a31388326a/writing_outputs/final/adc_oncology_pair_trade_report.pdf) |
| 61 | [U.S. Chocolate Market Valentine's Report](/examples/session_20260212_152520_8a0952fc765d/writing_outputs/final/US_Chocolate_Market_Report_Valentines_2026_v2.pdf) | Market Research | 4.0 weeks | $28,000 | 90 min | $50 | $27,950 | 560× | [Session](https://app.k-dense.ai/share/session_20260212_152520_8a0952fc765d) · [PDF](/examples/session_20260212_152520_8a0952fc765d/writing_outputs/final/US_Chocolate_Market_Report_Valentines_2026_v2.pdf) |
| 62 | [CD20×CD3 Bispecific Antibody Biomarker Discovery](/blog/dlbcl-biomarker-discovery-bispecific-antibodies) | Cancer Biology | 6.1 weeks | $42,700 | 3.7 hr | $100 | $42,600 | 427× | [Session](https://app.k-dense.ai/share/session_20260403_101458_7f548077d599) · [PDF](/examples/session_20260403_101458_7f548077d599/writing_outputs/final/biomarker_whitepaper_cd20xcd3.pdf) · [Blog](/blog/dlbcl-biomarker-discovery-bispecific-antibodies) |
| 63 | [PDAC Neoantigen Landscape & iNeST Strategy](/blog/pdac-neoantigen-landscape-inest-strategy) | Cancer Biology | 4.5 weeks | $31,500 | 4.0 hr | $100 | $31,400 | 315× | [Session](https://app.k-dense.ai/share/session_20260403_102916_6cb2547db0b9) · [PDF](/examples/session_20260403_102916_6cb2547db0b9/writing_outputs/final/PDAC_Neoantigen_Landscape_iNeST_Presentation.pdf) · [Blog](/blog/pdac-neoantigen-landscape-inest-strategy) |
| 64 | [CT-388 Obesity Drug Competitive Intelligence](/blog/ct388-competitive-intelligence-obesity) | Biotech Investment | 6.4 weeks | $76,800 | 5.0 hr | $150 | $76,650 | 512× | [Session](https://app.k-dense.ai/share/session_20260403_103155_c18ae3efc2c1) · [PDF](/examples/session_20260403_103155_c18ae3efc2c1/writing_outputs/final/CT388_Competitive_Intelligence_Report_2026.pdf) · [Blog](/blog/ct388-competitive-intelligence-obesity) |
| 65 | [ARPA-H Health Innovation Policy Report](/blog/arpa-h-policy-report-analysis) | Public Policy | 7.7 weeks | $61,600 | 4.0 hr | $185 | $61,415 | 333× | [Session](https://app.k-dense.ai/share/session_20260427_173439_56c05e499816) · [PDF](/examples/session_20260427_173439_56c05e499816/writing_outputs/final/ARPA-H_Policy_Report.pdf) · [Blog](/blog/arpa-h-policy-report-analysis) |
| 66 | [Appalachian Lithium Resource Techno-Economic Analysis](/examples/session_20260501_125918_14ec7ecb40b5/writing_outputs/final/Appalachian_Lithium_TEA_Report.pdf) | Climate Tech | 6.1 weeks | $54,900 | 4.5 hr | $120 | $54,780 | 458× | [Session](https://app.k-dense.ai/share/session_20260501_125918_14ec7ecb40b5) · [PDF](/examples/session_20260501_125918_14ec7ecb40b5/writing_outputs/final/Appalachian_Lithium_TEA_Report.pdf) |
| 67 | [Oil-Shock Macro Stress Test](/examples/session_20260505_134929_2f742528f8de/writing_outputs/final/oil_shock_macro_stress_test_2026Q2.pdf) | Macro Strategy | 3.7 weeks | $37,000 | 2.2 hr | $50 | $36,950 | 740× | [Session](https://app.k-dense.ai/share/session_20260505_134929_2f742528f8de) · [PDF](/examples/session_20260505_134929_2f742528f8de/writing_outputs/final/oil_shock_macro_stress_test_2026Q2.pdf) |

---

## Per-use-case detail (scope, evidence, breakdown)

Each entry below shows the work K-Dense Web actually produced, the estimated human-hour breakdown that drove the cost number, the K-Dense credit/wall-clock estimate, and links to the session.


### 1. Natural Products Chemical Space — Drug Discovery

**Links:** [Session](https://app.k-dense.ai/share/session_20251212_135154_dad1cf0d70ee) · [PDF](/examples/session_20251212_135154_dad1cf0d70ee/writing_outputs/final/NP_Antimicrobial_Screening_Manuscript.pdf) · [Blog](/blog/antimicrobial-drug-discovery-natural-products)

**Scope.** Filtered COCONUT (~716K NPs) to ~24.9K bacterial compounds, computed RDKit descriptors and drug-likeness filters, ran PCA/t-SNE/UMAP with K-Means and DBSCAN, prioritized candidates, and wrote a LaTeX manuscript with embedded figures.

**Evidence.** ~20 pages, 11 figures. Datasets: COCONUT Dec 2025 CSV; 715822 compounds filtered to 24910 bacterial NPs with RDKit-validated SMILES. Analyses: PCA t-SNE UMAP K-Means DBSCAN on molecular descriptors; QED Lipinski PAINS filtering; candidate shortlisting from chemical space clusters.

**Human estimate.** 232 hours × $175/hr = **$40,600** (5.8 full-time weeks). Breakdown: scoping/lit 24 h · data 56 h · analysis 72 h · figures 20 h · writing 50 h · review 8 h.

**K-Dense Web.** Heavy tier, ~140 credits ≈ **$100** at Plus rate, ~45 min wall clock.

**Savings.** $40,500 cost saved · 309× time compression · 406× cost multiplier.

_Notes: Repo has no committed PNG/PDF binaries; figures and page count taken from writing_outputs/SUMMARY.md and progress.md. Blog states ~45 min end-to-end runtime._


### 2. Macroeconomic Recession Prediction — Economics

**Links:** [Session](https://app.k-dense.ai/share/session_20251228_135128_cd3e46eedd35) · [PDF](/examples/session_20251228_135128_cd3e46eedd35/writing_outputs/final/recession_early_warning_paper.pdf)

**Scope.** Built recession early-warning models on 19 FRED indicators (1992–2024) with 205 engineered features using logit/probit benchmarks, Random Forest and XGBoost, plus PyMC Bayesian inference with uncertainty, compiled into an 18-page LaTeX paper.

**Evidence.** ~18 pages, 5 figures. Datasets: FRED macro series 1992–2024; 19 core indicators expanded to 205 features. Analyses: Probit logit baselines; Random Forest XGBoost; Bayesian regression PyMC; ROC and timeline visualizations; expanding-window validation.

**Human estimate.** 212 hours × $200/hr = **$42,400** (5.3 full-time weeks). Breakdown: scoping/lit 20 h · data 40 h · analysis 88 h · figures 10 h · writing 46 h · review 8 h.

**K-Dense Web.** Medium tier, ~70 credits ≈ **$50** at Plus rate, ~90 min wall clock.

**Savings.** $42,350 cost saved · 141× time compression · 848× cost multiplier.

_Notes: SUMMARY section title says six figures but enumerates five PNG deliverables; count follows the explicit list. No PNG files present in the workspace clone._


### 3. Longevity Gene Analysis — Genomics

**Links:** [Session](https://app.k-dense.ai/share/session_20251215_165944_a1f9981a1a45) · [PDF](/examples/session_20251215_165944_a1f9981a1a45/writing_outputs/final/longevity_translation_scorecard.pdf)

**Scope.** Synthesized longevity genetics evidence into a translation scorecard with GWAS counts, STRING network, druggability heatmap, and funnel visualization, formatted as a 16-page PDF report.

**Evidence.** ~16 pages, 5 figures. Datasets: Ortholog mapping and GWAS-derived longevity gene lists summarized in session manifests and dossiers. Analyses: Network visualization STRING context; druggability matrix by clinical phase; comparative bar and Venn summaries for translation gaps.

**Human estimate.** 188 hours × $175/hr = **$32,900** (4.7 full-time weeks). Breakdown: scoping/lit 16 h · data 52 h · analysis 60 h · figures 12 h · writing 40 h · review 6 h.

**K-Dense Web.** Medium tier, ~60 credits ≈ **$50** at Plus rate, ~75 min wall clock.

**Savings.** $32,850 cost saved · 150× time compression · 658× cost multiplier.


### 4. JWST Exoplanet Target Prioritization — Astronomy

**Links:** [Session](https://app.k-dense.ai/share/session_20251214_135820_a55c35bcb856) · [PDF](/examples/session_20251214_135820_a55c35bcb856/writing_outputs/final/JWST_Biosignature_Priority_Framework.pdf)

**Scope.** Ranked transiting habitable-zone targets for JWST biosignature follow-up using observability habitat and stellar activity metrics, statistical comparisons, and scheduling visuals in a 22-page framework document.

**Evidence.** ~22 pages, 8 figures. Datasets: Curated exoplanet target table with TSM ESM and JWST visibility-related metrics per README and manifest. Analyses: Metric distributions; activity penalty sensitivity; hypothesis tests M dwarf vs non-M-dwarf; habitability vs observability trade space; 2-year Gantt-style schedule.

**Human estimate.** 200 hours × $200/hr = **$40,000** (5.0 full-time weeks). Breakdown: scoping/lit 24 h · data 44 h · analysis 52 h · figures 16 h · writing 56 h · review 8 h.

**K-Dense Web.** Heavy tier, ~115 credits ≈ **$100** at Plus rate, ~2.5 hr wall clock.

**Savings.** $39,900 cost saved · 80× time compression · 400× cost multiplier.


### 5. Quantum Chemistry VQE Benchmarking — Quantum Chemistry

**Links:** [Session](https://app.k-dense.ai/share/session_20251212_162938_43788a48ae9a) · [PDF](/examples/session_20251212_162938_43788a48ae9a/writing_outputs/final/vqe_benchmarking_paper.pdf)

**Scope.** Benchmarked VQE on small molecules with PennyLane simulations comparing ansätze scaling and noise robustness, reported in a 15-page manuscript with workflow schematic plus four analysis figures.

**Evidence.** ~15 pages, 5 figures. Datasets: Stored molecular benchmark pickles with PES and resource metrics for H2 LiH BeH2 H2O per session README. Analyses: VQE energy surfaces vs FCI reference; parameter and qubit scaling; depolarizing noise thresholds; HardwareEfficient vs UCCSD-style ansatz comparison.

**Human estimate.** 188 hours × $225/hr = **$42,300** (4.7 full-time weeks). Breakdown: scoping/lit 20 h · data 36 h · analysis 76 h · figures 12 h · writing 38 h · review 6 h.

**K-Dense Web.** Medium tier, ~85 credits ≈ **$50** at Plus rate, ~2.0 hr wall clock.

**Savings.** $42,250 cost saved · 94× time compression · 846× cost multiplier.


### 6. Doxorubicin Mechanism of Action — Cancer Biology

**Links:** [Session](https://app.k-dense.ai/share/session_20251217_123457_77eda5efa279)

**Scope.** RNA-seq workflow on GSE244574 MCF7 vehicle vs doxorubicin with QC DE GSEA ORA connectivity mining integrated dashboard and pathway schematic producing final_paper.md (~14k words; no PDF checked in).

**Evidence.** ~24 pages, 9 figures. Datasets: GEO GSE244574 16 samples 8 control 8 treated; gene lists and enrichment outputs in workflow results. Analyses: PCA QC differential expression GSEA ORA CLUE-style connectivity MoA dashboard mechanism schematic.

**Human estimate.** 216 hours × $175/hr = **$37,800** (5.4 full-time weeks). Breakdown: scoping/lit 16 h · data 44 h · analysis 68 h · figures 18 h · writing 60 h · review 10 h.

**K-Dense Web.** Heavy tier, ~105 credits ≈ **$50** at Plus rate, ~2.0 hr wall clock.

**Savings.** $37,750 cost saved · 108× time compression · 756× cost multiplier.

_Notes: pdfPath absent in site data; page estimate from README word-count guidance (~14000 words) using ~600 words per formatted manuscript page. Figure count matches nine PNG paths in manifest plus schematic also saved as PDF._


### 7. ECG Stress Detection Pipeline — Biosignals

**Links:** [Session](https://app.k-dense.ai/share/session_20251212_210544_328dfde1326e) · [PDF](/examples/session_20251212_210544_328dfde1326e/writing_outputs/final/hrv_stress_detection_manuscript.pdf)

**Scope.** End-to-end HRV feature pipeline from raw ECG through exploratory plots classifier tuning and comparison culminating in an 18-page stress-detection manuscript plus infographic.

**Evidence.** ~18 pages, 9 figures. Datasets: Processed ECG-derived records and engineered HRV features referenced across README step summaries. Analyses: Signal QC correlation heatmaps distribution plots confusion matrix ROC feature importance multi-model comparison supervised classifier.

**Human estimate.** 208 hours × $175/hr = **$36,400** (5.2 full-time weeks). Breakdown: scoping/lit 16 h · data 48 h · analysis 76 h · figures 14 h · writing 46 h · review 8 h.

**K-Dense Web.** Heavy tier, ~95 credits ≈ **$50** at Plus rate, ~2.0 hr wall clock.

**Savings.** $36,350 cost saved · 104× time compression · 728× cost multiplier.

_Notes: progress.md reports nine figures total infographic plus eight analysis panels._


### 8. Lycopene Biosynthesis Optimization — Synthetic Biology

**Links:** [Session](https://app.k-dense.ai/share/session_20251215_170752_24b906eb7e01) · [PDF](/examples/session_20251215_170752_24b906eb7e01/writing_outputs/final/manuscript.pdf)

**Scope.** Kinetic modeling of heterologous lycopene pathway enzymes bottleneck scoring and multi-objective GA optimization with four integrated manuscript figures and AI framework schematic.

**Evidence.** ~17 pages, 4 figures. Datasets: Parameterized enzyme kinetic sets pathway stoichiometry and optimization budgets described in README paper.md. Analyses: ODE-style kinetic simulation bottleneck scoring genetic algorithm Pareto fronts across expression budgets.

**Human estimate.** 164 hours × $175/hr = **$28,700** (4.1 full-time weeks). Breakdown: scoping/lit 16 h · data 40 h · analysis 52 h · figures 8 h · writing 42 h · review 6 h.

**K-Dense Web.** Medium tier, ~65 credits ≈ **$50** at Plus rate, ~90 min wall clock.

**Savings.** $28,650 cost saved · 109× time compression · 574× cost multiplier.

_Notes: SUMMARY lists four analysis figures plus framework schematic counted here as part of the four integrated outputs referenced in progress integration checklist._


### 9. Epigenetic Clock Drug Discovery — Epigenetics

**Links:** [Session](https://app.k-dense.ai/share/session_20251215_170449_b3d275ad14ce) · [PDF](/examples/session_20251215_170449_b3d275ad14ce/writing_outputs/final/epigenetic_aging_mechanisms.pdf)

**Scope.** Integrated dual GEO methylation cohorts for aging-clock master regulators TF binding overlap validation charts and AI mechanistic schematic compiled into a 15-page LaTeX PDF.

**Evidence.** ~15 pages, 6 figures. Datasets: GSE40279 GSE87571 downloads summarized under workflow data JSON summaries. Analyses: DMP prioritization master regulator ranking drug-target overlays genomic distribution TF overlap validation panels mechanistic figure.

**Human estimate.** 204 hours × $175/hr = **$35,700** (5.1 full-time weeks). Breakdown: scoping/lit 16 h · data 56 h · analysis 72 h · figures 14 h · writing 38 h · review 6 h.

**K-Dense Web.** Medium tier, ~85 credits ≈ **$50** at Plus rate, ~2.0 hr wall clock.

**Savings.** $35,650 cost saved · 102× time compression · 714× cost multiplier.


### 10. GLP-1 Drug Patent Landscape — Pharma/IP

**Links:** [Session](https://app.k-dense.ai/share/session_20251214_114340_87efd8b52064) · [PDF](/examples/session_20251214_114340_87efd8b52064/writing_outputs/final/patent_vulnerability_report.pdf)

**Scope.** Linked USPTO patents to GLP-1 drugs via ChemBL temporal inversion vulnerability motifs visualized with velocity network lag charts plus schematic in a 15-page patent vulnerability PDF.

**Evidence.** ~15 pages, 4 figures. Datasets: USPTO patent extracts ChemBL compound linkage tables paper drug mappings JSON. Analyses: Publication velocity trends co-occurrence network Jaccard weights FDA lag bars temporal inversion schematic narrative.

**Human estimate.** 156 hours × $175/hr = **$27,300** (3.9 full-time weeks). Breakdown: scoping/lit 12 h · data 48 h · analysis 44 h · figures 8 h · writing 38 h · review 6 h.

**K-Dense Web.** Medium tier, ~55 credits ≈ **$50** at Plus rate, ~75 min wall clock.

**Savings.** $27,250 cost saved · 125× time compression · 546× cost multiplier.


### 11. Biodegradable Polymer Safety Analysis — Biomedical Materials

**Links:** [Session](https://app.k-dense.ai/share/session_20251212_210923_d71aa9ce43f6) · [PDF](/examples/session_20251212_210923_d71aa9ce43f6/writing_outputs/final/manuscript.pdf)

**Scope.** Statistically summarized MAUDE adverse events for PLLA PLGA devices correlating failure modes with material properties adding workflow schematic and decision flowchart in an 18-page manuscript.

**Evidence.** ~18 pages, 5 figures. Datasets: FDA MAUDE-derived adverse event records stratified by polymer type per README. Analyses: Frequency tables correlation heatmap with significance grouped bar charts decision flowchart for selection risk.

**Human estimate.** 168 hours × $175/hr = **$29,400** (4.2 full-time weeks). Breakdown: scoping/lit 16 h · data 44 h · analysis 42 h · figures 10 h · writing 46 h · review 8 h.

**K-Dense Web.** Medium tier, ~70 credits ≈ **$50** at Plus rate, ~90 min wall clock.

**Savings.** $29,350 cost saved · 112× time compression · 588× cost multiplier.


### 12. ICD Adverse Events Analysis — Medical Devices

**Links:** [Session](https://app.k-dense.ai/share/session_20251212_210203_5e688fa2ddca) · [PDF](/examples/session_20251212_210203_5e688fa2ddca/writing_outputs/final/ICD_Adverse_Events_Technical_Report.pdf)

**Scope.** Analyzed 10,000 ICD MAUDE adverse events (Apr–Jul 2020) with keyword categorization, LDA/NMF topic modeling, statistical comparisons across manufacturers, and publication-style reporting.

**Evidence.** ~34 pages, 7 figures. Datasets: openFDA MAUDE ICD events (10000 records, Apr–Jul 2020); processed CSVs and NLP topic assignments. Analyses: Keyword failure-mode taxonomy; LDA/NMF topic models; chi-square/Fisher tests with FDR; pairwise ORs; manufacturer–failure network visualization.

**Human estimate.** 228 hours × $175/hr = **$39,900** (5.7 full-time weeks). Breakdown: scoping/lit 24 h · data 32 h · analysis 64 h · figures 12 h · writing 84 h · review 12 h.

**K-Dense Web.** Heavy tier, ~150 credits ≈ **$99** at Plus rate, ~3.0 hr wall clock.

**Savings.** $39,800 cost saved · 76× time compression · 403× cost multiplier.

_Notes: 34 pages via LaTeX build log; seven figures listed in SUMMARY including AI failure taxonomy (writing_outputs); core report embeds six vector figures in TeX._


### 13. Nanoparticle Protein Corona Prediction — Nanomedicine

**Links:** [Session](https://app.k-dense.ai/share/session_20251214_140128_b84388b8e99e)

**Scope.** Built a corona modeling pipeline on 144 protein–NP associations with EDA, LOGO ML, AlphaFold-style structural validation, a CLI calculator, and a 10-slide results deck.

**Evidence.** ~10 pages, 7 figures. Datasets: Unified corona CSV (144 rows × 17 enriched features); structural metrics for 9 proteins; literature-grounded synthetic curation per README. Analyses: LOGO cross-validated ML abundance models; RDKit ligand descriptors; structural EHSA/GRAVY validation; corona_design_calculator CLI.

**Human estimate.** 144 hours × $175/hr = **$25,200** (3.6 full-time weeks). Breakdown: scoping/lit 24 h · data 28 h · analysis 64 h · figures 12 h · writing 12 h · review 4 h.

**K-Dense Web.** Heavy tier, ~70 credits ≈ **$46** at Plus rate, ~80 min wall clock.

**Savings.** $25,150 cost saved · 108× time compression · 548× cost multiplier.

_Notes: pdf_pages set to slide count (manifest PPTX); wall clock inferred from completed_steps timestamps (~80 min)._


### 14. MS Fragmentation Rule Mining — Mass Spectrometry

**Links:** [Session](https://app.k-dense.ai/share/session_20251215_165434_bcb1a47df161) · [PDF](/examples/session_20251215_165434_bcb1a47df161/writing_outputs/final/fragmentation_rules_paper.pdf)

**Scope.** End-to-end fragmentation rule discovery with validation, bipartite rule–metabolite network visualization, and a 15-page LaTeX paper with references.

**Evidence.** ~15 pages, 2 figures. Datasets: Simulated/curated fragmentation inputs per data pipeline (MANIFEST); processed rule statistics and network GEXF. Analyses: Rule mining and statistical validation pipeline; bipartite network layout/export; rule application tooling.

**Human estimate.** 156 hours × $175/hr = **$27,300** (3.9 full-time weeks). Breakdown: scoping/lit 20 h · data 40 h · analysis 52 h · figures 4 h · writing 36 h · review 4 h.

**K-Dense Web.** Medium tier, ~60 credits ≈ **$40** at Plus rate, ~90 min wall clock.

**Savings.** $27,250 cost saved · 104× time compression · 683× cost multiplier.

_Notes: Figure count from SUMMARY (workflow schematic + network); PDF page count from v1_draft build log._


### 15. Wearable Seizure Prediction — Neuroscience

**Links:** [Session](https://app.k-dense.ai/share/session_20251214_140801_5fd244475cff) · [PDF](/examples/session_20251214_140801_5fd244475cff/writing_outputs/final/manuscript.pdf)

**Scope.** Multi-dataset wearable biosignal pipeline (CHB-MIT/DEAP) with feature engineering, classifiers, phase–amplitude coupling analysis, and a 22-page manuscript.

**Evidence.** ~22 pages, 5 figures. Datasets: Public EEG/HRV-style corpora per README (CHB-MIT chunks, DEAP references); merged processed matrices in data/processed. Analyses: Preprocessing and feature extraction; seizure vs baseline classifiers; heart–brain coupling PAC module; performance and feasibility comparisons.

**Human estimate.** 212 hours × $175/hr = **$37,100** (5.3 full-time weeks). Breakdown: scoping/lit 24 h · data 40 h · analysis 72 h · figures 12 h · writing 56 h · review 8 h.

**K-Dense Web.** Heavy tier, ~130 credits ≈ **$86** at Plus rate, ~2.5 hr wall clock.

**Savings.** $37,000 cost saved · 85× time compression · 431× cost multiplier.

_Notes: Five primary figures per SUMMARY (workflow schematic plus four analysis figures); page count from v1_manuscript.log._


### 16. Air Quality Health Impact Analysis — Environment

**Links:** [Session](https://app.k-dense.ai/share/session_20251212_124112_f9f3e9939b7d)

**Scope.** Merged EPA daily AQI with census populations (325k+ rows), trained multi-class air-quality classifiers, generated seven EDA/ML figures, and produced literature and modeling memos.

**Evidence.** ~10 pages, 7 figures. Datasets: daily_aqi_by_county_2023 (325399 rows); NST-EST2023 population; merged parquet 323k+ modeling rows. Analyses: Literature synthesis (10 papers); XGBoost 3-way AQI category classifier with lag features; accuracy ~0.80 per model_report.txt.

**Human estimate.** 144 hours × $200/hr = **$28,800** (3.6 full-time weeks). Breakdown: scoping/lit 24 h · data 40 h · analysis 48 h · figures 8 h · writing 20 h · review 4 h.

**K-Dense Web.** Medium tier, ~55 credits ≈ **$36** at Plus rate, ~90 min wall clock.

**Savings.** $28,750 cost saved · 96× time compression · 800× cost multiplier.

_Notes: No LaTeX PDF in tree; pdf_pages estimated from literature_synthesis + model_report + README narrative (~10 page equivalent); seven PNG paths from manifest.json._


### 17. LC-MS Ion Suppression Prediction — Metabolomics

**Links:** [Session](https://app.k-dense.ai/share/session_20251215_164703_dc8bf4ef0471) · [PDF](/examples/session_20251215_164703_dc8bf4ef0471/writing_outputs/final/ion_suppression_paper.pdf)

**Scope.** Metabolomics Workbench multi-study pull LC-MS features, tuned ML for ion suppression factor prediction, SHAP interpretability, and a 24-page manuscript with workflow and EDA figures.

**Evidence.** ~24 pages, 3 figures. Datasets: Multiple MW study metadata and metabolite tables (ST* studies); merged modeling frame per README. Analyses: Feature engineering on LC-MS inputs; gradient boosting-style winner with CV; SHAP summaries; peer-reviewed draft per PEER_REVIEW.md.

**Human estimate.** 216 hours × $175/hr = **$37,800** (5.4 full-time weeks). Breakdown: scoping/lit 24 h · data 48 h · analysis 68 h · figures 8 h · writing 60 h · review 8 h.

**K-Dense Web.** Heavy tier, ~120 credits ≈ **$79** at Plus rate, ~3.0 hr wall clock.

**Savings.** $37,700 cost saved · 72× time compression · 478× cost multiplier.

_Notes: Three final science figures in manuscript (workflow schematic, multi-panel EDA, SHAP) per SUMMARY; pages from v1_draft.log._


### 18. Agent Skills vs MCP Architecture — AI/Technical

**Links:** [Session](https://app.k-dense.ai/share/session_20251231_185247_6dce8fea6faa) · [PDF](/examples/session_20251231_185247_6dce8fea6faa/writing_outputs/final/agent_skills_vs_mcp_report.pdf)

**Scope.** Architecture comparison report contrasting MCP vs file-based Agent Skills with diagrams, data-flow graphics, and a ~22-page technical PDF plus Markdown source.

**Evidence.** ~22 pages, 5 figures. Datasets: Reference architecture and protocol docs synthesized in prose (no empirical dataset). Analyses: Comparative systems analysis (protocol primitives, lifecycle, security, extensibility); tabular scoring in figures.

**Human estimate.** 136 hours × $200/hr = **$27,200** (3.4 full-time weeks). Breakdown: scoping/lit 20 h · data 8 h · analysis 36 h · figures 8 h · writing 56 h · review 8 h.

**K-Dense Web.** Medium tier, ~65 credits ≈ **$43** at Plus rate, ~105 min wall clock.

**Savings.** $27,150 cost saved · 78× time compression · 633× cost multiplier.

_Notes: Page count from v1_draft.log; five diagram figures per SUMMARY.md table._


### 19. TMT Proteomics Spike-in Validation — Proteomics

**Links:** [Session](https://app.k-dense.ai/share/session_20251212_165231_ad0681ba0d19) · [PDF](/examples/session_20251212_165231_ad0681ba0d19/writing_outputs/final/Grant_Report.pdf)

**Scope.** PXD000001 TMT spike-in reanalysis with QC normalization, rank/ratio diagnostics, and grant-style write-up (16-page LaTeX) plus slide-ready exports.

**Evidence.** ~16 pages, 4 figures. Datasets: PXD000001 mzTab quantification (~2k rows) plus derived spike-in validation stats JSON. Analyses: Intensity normalization and QC PCA; spike-in expected vs observed ratios; variance-ranked protein QC.

**Human estimate.** 152 hours × $175/hr = **$26,600** (3.8 full-time weeks). Breakdown: scoping/lit 20 h · data 36 h · analysis 40 h · figures 8 h · writing 40 h · review 8 h.

**K-Dense Web.** Medium tier, ~60 credits ≈ **$40** at Plus rate, ~75 min wall clock.

**Savings.** $26,550 cost saved · 122× time compression · 665× cost multiplier.

_Notes: Four manuscript figures per TEX includes plus SUMMARY; page count from v1_draft.log._


### 20. T2-Low Asthma Endotype Discovery — Genomics

**Links:** [Session](https://app.k-dense.ai/share/session_20251216_063845_10f14a0ac5fe) · [PDF](/examples/session_20251216_063845_10f14a0ac5fe/writing_outputs/final/Asthma_Endotyping_Presentation.pdf)

**Scope.** GEO-based asthma endotyping: QC, DEG, multi-database enrichment, patient stratification and clustering, 12 analysis figures, AI graphical abstract, and an 18-slide PDF deck.

**Evidence.** ~18 pages, 13 figures. Datasets: GEO asthma transcriptomic archive processed per data/README (HDF5 PKL pipeline); significant gene lists and pathway scores. Analyses: Differential expression; GO/KEGG/Reactome enrichment; pathway-score stratification; k-means endotype clustering with silhouettes.

**Human estimate.** 196 hours × $175/hr = **$34,300** (4.9 full-time weeks). Breakdown: scoping/lit 28 h · data 40 h · analysis 80 h · figures 24 h · writing 20 h · review 4 h.

**K-Dense Web.** Heavy tier, ~155 credits ≈ **$102** at Plus rate, ~3.0 hr wall clock.

**Savings.** $34,200 cost saved · 65× time compression · 336× cost multiplier.

_Notes: Figure count: 12 stepwise PNGs in MANIFEST.md plus graphical_abstract.png per writing_outputs/SUMMARY; session duration ~3 h per README; writing used 1 hr/slide rule for 18 slides._


### 21. Caribbean CBI Tax Strategy — Finance/Legal

**Links:** [Session](https://app.k-dense.ai/share/session_20251217_072723_efc7231a0106) · [PDF](/examples/session_20251217_072723_efc7231a0106/writing_outputs/final/executive_decision_report.pdf)

**Scope.** Six-step Caribbean CBI research pack (costs, tax, logistics, legal CSME, roadmap, synthesis) distilled into a 20-page executive PDF with ROI and Gantt figures.

**Evidence.** ~20 pages, 3 figures. Datasets: Structured memos in data/research (step1–6 markdown) with quantitative CBI cost and ROI assumptions. Analyses: Hybrid operational structure framing; 10-year ROI scenario; phased implementation and dependency Gantt.

**Human estimate.** 164 hours × $250/hr = **$41,000** (4.1 full-time weeks). Breakdown: scoping/lit 32 h · data 24 h · analysis 44 h · figures 4 h · writing 52 h · review 8 h.

**K-Dense Web.** Medium tier, ~55 credits ≈ **$36** at Plus rate, ~90 min wall clock.

**Savings.** $40,950 cost saved · 109× time compression · 1.1K× cost multiplier.

_Notes: Three AI-generated strategic charts per SUMMARY; page count from v1_executive_report.log._


### 22. HNSCC Treatment Response Biomarkers — Cancer Biology

**Links:** [Session](https://app.k-dense.ai/share/session_20251217_122112_5804b26745cf)

**Scope.** GSE65858 HNSCC cohort (270 samples) DEA plus optimized Random Forest response model (AUC ~0.76) with six diagnostic figures and a full markdown research paper (no PDF in tree).

**Evidence.** ~14 pages, 6 figures. Datasets: GEO GSE65858 expression matrix (31330 probes × 270 samples) with CR/PD labels. Analyses: Differential expression; top-20 gene signature; RandomForestClassifier with grid search CV; test metrics and feature importances.

**Human estimate.** 160 hours × $175/hr = **$28,000** (4.0 full-time weeks). Breakdown: scoping/lit 24 h · data 32 h · analysis 48 h · figures 12 h · writing 36 h · review 8 h.

**K-Dense Web.** Medium tier, ~65 credits ≈ **$43** at Plus rate, ~105 min wall clock.

**Savings.** $27,950 cost saved · 91× time compression · 651× cost multiplier.

_Notes: pdf_pages estimated from workflow/research_paper.md length (~14 equivalent pages); six PNGs listed in manifest excluding PDF duplicates._


### 23. Tox21 Liver Toxicity Prediction — Toxicology

**Links:** [Session](https://app.k-dense.ai/share/session_20251217_122902_9ad327069ee6) · [PDF](/examples/session_20251217_122902_9ad327069ee6/writing_outputs/final/predictive_toxicology_paper.pdf)

**Scope.** Built an XGBoost model on Tox21 SR-MMP data with Morgan fingerprints, predicted five leads, and documented SHAP-driven structural interpretation in a cited LaTeX paper.

**Evidence.** ~13 pages, 6 figures. Datasets: Tox21 SR-MMP via DeepChem (6258 compounds); five synthetic leads; featurized CSVs. Analyses: 2048-bit Morgan ECFP4; XGBoost binary classifier with 5-fold CV; SHAP TreeExplainer on LEAD_005.

**Human estimate.** 164 hours × $175/hr = **$28,700** (4.1 full-time weeks). Breakdown: scoping/lit 24 h · data 32 h · analysis 56 h · figures 14 h · writing 32 h · review 5 h.

**K-Dense Web.** Medium tier, ~55 credits ≈ **$50** at Plus rate, ~75 min wall clock.

**Savings.** $28,650 cost saved · 131× time compression · 574× cost multiplier.

_Notes: Figure count and page count from writing_outputs/SUMMARY.md; repo has review logs but no committed PNGs._


### 24. US Vineyard Location Analysis — Agriculture

**Links:** [Session](https://app.k-dense.ai/share/session_20251218_143137_2f4bd2ba43ea) · [PDF](/examples/session_20251218_143137_2f4bd2ba43ea/writing_outputs/final/vineyard_location_report.pdf)

**Scope.** Multi-step regional research on Paso Robles, Sierra Foothills, and Texas Hill Country culminating in a weighted scoring matrix, risks, and five-year roadmap for a vineyard-plus-glamping investment.

**Evidence.** ~19 pages, 3 figures. Datasets: Qualitative/climate and regulatory research memos in research/*.md; economic scenarios in prose. Analyses: Weighted scoring across terroir, versatility, business ecosystem, economics, and quantified regional risks.

**Human estimate.** 148 hours × $175/hr = **$25,900** (3.7 full-time weeks). Breakdown: scoping/lit 36 h · data 28 h · analysis 24 h · figures 6 h · writing 48 h · review 8 h.

**K-Dense Web.** Medium tier, ~55 credits ≈ **$50** at Plus rate, ~90 min wall clock.

**Savings.** $25,850 cost saved · 99× time compression · 518× cost multiplier.

_Notes: Figures listed in writing_outputs/SUMMARY.md; only *_review_log.json appears under writing_outputs/figures/ in the repo snapshot._


### 25. Aravali Mountains Impact Prediction — Environment

**Links:** [Session](https://app.k-dense.ai/share/session_20251219_144034_6ca0fee97d6e) · [PDF](/examples/session_20251219_144034_6ca0fee97d6e/writing_outputs/final/aravali_crisis_research_paper.pdf)

**Scope.** Policy-grounded scenario modeling with Open-Meteo climate series plus synthetic AQI/groundwater trends, linear baselines, four tampering scenarios to 2050, and a long LaTeX policy paper.

**Evidence.** ~36 pages, 7 figures. Datasets: 18970 daily climate rows (Gurugram, Alwar); 9485 synthetic AQI days; 1560 monthly groundwater rows across five locations. Analyses: Linear regression BAU projections; scenario modifiers for temperature, PM2.5, and groundwater to 2050.

**Human estimate.** 248 hours × $200/hr = **$49,600** (6.2 full-time weeks). Breakdown: scoping/lit 32 h · data 48 h · analysis 44 h · figures 18 h · writing 92 h · review 14 h.

**K-Dense Web.** Mega tier, ~190 credits ≈ **$150** at Plus rate, ~6.0 hr wall clock.

**Savings.** $49,450 cost saved · 41× time compression · 331× cost multiplier.

_Notes: Page count follows SUMMARY.md Version 3 (36 pages); figures include three baseline and three scenario comparisons plus green-wall schematic per SUMMARY files listing._


### 26. Indian Vegetarian Diet Optimization — Nutrition

**Links:** [Session](https://app.k-dense.ai/share/session_20251221_093619_3390c5c3bed6) · [PDF](/examples/session_20251221_093619_3390c5c3bed6/writing_outputs/final/india_vegetarian_diet_plan.pdf)

**Scope.** Compiled ICMR-aligned vegetarian nutrition tables, optimized multi-day meal plans by activity level, generated comparative charts, and produced a multi-reference LaTeX report plus slide deck.

**Evidence.** ~16 pages, 9 figures. Datasets: 50-item indian_food_nutrition.csv scaling to expanded CSVs in SUMMARY; meal_plan_* and daily_summary_* CSVs. Analyses: Greedy/constrained meal assembly with portion multipliers across Sedentary, Moderate, and Active profiles.

**Human estimate.** 172 hours × $175/hr = **$30,100** (4.3 full-time weeks). Breakdown: scoping/lit 24 h · data 28 h · analysis 36 h · figures 14 h · writing 62 h · review 10 h.

**K-Dense Web.** Heavy tier, ~135 credits ≈ **$100** at Plus rate, ~2.0 hr wall clock.

**Savings.** $30,000 cost saved · 86× time compression · 301× cost multiplier.

_Notes: SUMMARY cites a 16-page urban_diet_report.pdf and 21 Beamer slides while manifest lists 11-slide Nano Banana PDF; figure PNGs enumerated in manifest include nine PNGs ignoring duplicate PDFs._


### 27. East vs West Cultural Analysis — Social Science

**Links:** [Session](https://app.k-dense.ai/share/session_20251223_092351_e1527e79a0d2) · [PDF](/examples/session_20251223_092351_e1527e79a0d2/writing_outputs/final/presentation.pdf)

**Scope.** Merged Hofstede and World Happiness indicators for 41 countries, ran t-tests/correlations highlighting India, produced six analytic plots plus workflow schematic, and packaged a 14-slide PDF deck.

**Evidence.** ~14 pages, 7 figures. Datasets: hofstede_raw.csv and happiness_raw.csv with merged processed_data.csv (41 countries). Analyses: Independent samples t-tests, Pearson correlations, Cohen's d; India spotlight analyses.

**Human estimate.** 92 hours × $175/hr = **$16,100** (2.3 full-time weeks). Breakdown: scoping/lit 20 h · data 16 h · analysis 24 h · figures 16 h · writing 14 h · review 2 h.

**K-Dense Web.** Light tier, ~30 credits ≈ **$50** at Plus rate, ~45 min wall clock.

**Savings.** $16,050 cost saved · 123× time compression · 322× cost multiplier.

_Notes: Writing hours use 1 hr/slide for the PDF deck; pdf_pages treated as slide count per SUMMARY.md._


### 28. Renal Cancer Drug Response Stratification — Cancer Biology

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_080443_4cf56e41bc8e)

**Scope.** Merged ArchS4 RCC/ccRCC RNA-seq, PyDESeq2 DEA, corrected TF enrichment, pathway scoring for four drug classes, responder heuristic stratification, DSigDB repurposing screen, and Markdown manuscript.

**Evidence.** ~30 pages, 10 figures. Datasets: 213 QC-pass samples × ~53k genes; drug knowledge base and pathway scores for 97 ccRCC samples. Analyses: PyDESeq2; GSEApy Enrichr TF databases; Scanpy gene-set scoring; Wilcoxon DE for non-responders; DSigDB enrichment.

**Human estimate.** 288 hours × $175/hr = **$50,400** (7.2 full-time weeks). Breakdown: scoping/lit 28 h · data 56 h · analysis 116 h · figures 22 h · writing 60 h · review 10 h.

**K-Dense Web.** Heavy tier, ~145 credits ≈ **$100** at Plus rate, ~4.0 hr wall clock.

**Savings.** $50,300 cost saved · 72× time compression · 504× cost multiplier.

_Notes: No PDF path in the catalog; pdf_pages estimated from manifest word_count ~15000 at ~500 words/page; ten PNGs counted from manifest ignoring paired PDFs._


### 29. Indian State Emigration Patterns — Demographics

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_091022_74d424b00564) · [PDF](/examples/session_20251224_091022_74d424b00564/writing_outputs/final/Indian_Expat_Analysis_Presentation.pdf)

**Scope.** Harmonized state-level emigration, economic, social, and diaspora panels (2010–2023), regressions and correlations, eleven matplotlib figures, and a sixteen-slide branded presentation PDF.

**Evidence.** ~16 pages, 11 figures. Datasets: Four raw CSVs merged into unified_dataset.csv and state_summary.csv (~280 state-year observations). Analyses: Correlation matrices, simple and multiple regressions, variance decomposition narrative (78.4 percent explained).

**Human estimate.** 124 hours × $175/hr = **$21,700** (3.1 full-time weeks). Breakdown: scoping/lit 16 h · data 40 h · analysis 36 h · figures 18 h · writing 16 h · review 2 h.

**K-Dense Web.** Medium tier, ~50 credits ≈ **$50** at Plus rate, ~75 min wall clock.

**Savings.** $21,650 cost saved · 99× time compression · 434× cost multiplier.

_Notes: Slide-style deliverable; pdf_pages equals slide count from writing_outputs/PRESENTATION_SUMMARY.md._


### 30. 50m Concrete Dome Structural Analysis — Engineering

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_102759_927c1cef200e) · [PDF](/examples/session_20251224_102759_927c1cef200e/writing_outputs/final/50m_dome_technical_report.pdf)

**Scope.** Membrane-theory structural checks, AS 1170 load combinations, quantity/cost ROM for Sydney, pneumatic formwork recommendation, and a 44-page standards-heavy LaTeX technical report.

**Evidence.** ~44 pages, 10 figures. Datasets: Loads, stresses, quantities, and CSV summaries generated by Python workflows. Analyses: Thin-shell membrane stress distributions; ROM costing ±30 percent with pneumatic vs traditional comparison.

**Human estimate.** 232 hours × $200/hr = **$46,400** (5.8 full-time weeks). Breakdown: scoping/lit 24 h · data 16 h · analysis 48 h · figures 16 h · writing 112 h · review 16 h.

**K-Dense Web.** Heavy tier, ~125 credits ≈ **$100** at Plus rate, ~3.0 hr wall clock.

**Savings.** $46,300 cost saved · 77× time compression · 464× cost multiplier.

_Notes: Nine quantitative PNGs in manifest plus AI construction schematic noted in writing_outputs/SUMMARY.md._


### 31. India Urban Air Pollution Tipping Points — Environment

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_200424_4a6cc33ed0fd) · [PDF](/examples/session_20251224_200424_4a6cc33ed0fd/writing_outputs/final/india_pollution_research_paper.pdf)

**Scope.** World Bank plus synthetic daily panels for four metros, Mann-Kendall/Pettitt/PELT tipping-point diagnostics, Holt-Winters forecasts, comparative synchronization analysis, 30-page enhanced research PDF with extra schematics, and eighteen-slide deck.

**Evidence.** ~30 pages, 8 figures. Datasets: Cleaned World Bank PM2.5; 43836 synthetic daily rows for four cities; monthly/yearly aggregates. Analyses: Mann-Kendall, Pettitt, ruptures PELT, STL decomposition, Holt-Winters forecasting, Pearson synchronization matrix.

**Human estimate.** 256 hours × $200/hr = **$51,200** (6.4 full-time weeks). Breakdown: scoping/lit 28 h · data 48 h · analysis 56 h · figures 18 h · writing 94 h · review 14 h.

**K-Dense Web.** Mega tier, ~200 credits ≈ **$150** at Plus rate, ~5.0 hr wall clock.

**Savings.** $51,050 cost saved · 51× time compression · 341× cost multiplier.

_Notes: Research PDF page count from ENHANCEMENT_SUMMARY_v3.md; eight figures after adding causal pathway and policy timeline (six baseline analytics plus two schematics); slides add eighteen slide-hours in writing breakdown._


### 32. India vs China AI Research Analysis — Scientometrics

**Links:** [Session](https://app.k-dense.ai/share/session_20251226_095001_6ed1c15df75b) · [PDF](/examples/session_20251226_095001_6ed1c15df75b/writing_outputs/final/manuscript.pdf)

**Scope.** OpenAlex bibliometrics across baseline vs GenAI windows, productivity/impact/collaboration metrics with statistical tests, literature synthesis, corrected true-count figures, and a twenty-one-page manuscript with twenty-eight references.

**Evidence.** ~21 pages, 7 figures. Datasets: 40k sampled publication JSON chunks plus true_annual_counts.csv correction table. Analyses: t-tests, Mann-Whitney U, Cohen's d; counterfactual missed-publication estimate; figure regeneration on true counts.

**Human estimate.** 196 hours × $200/hr = **$39,200** (4.9 full-time weeks). Breakdown: scoping/lit 28 h · data 40 h · analysis 52 h · figures 16 h · writing 54 h · review 8 h.

**K-Dense Web.** Heavy tier, ~120 credits ≈ **$100** at Plus rate, ~2.5 hr wall clock.

**Savings.** $39,100 cost saved · 78× time compression · 392× cost multiplier.

_Notes: Figure count includes graphical abstract plus five main panels plus corrected ratio figure described in manifest visualizations._


### 33. India Sacred Tree Tourism Strategy — Tourism/Economics

**Links:** [Session](https://app.k-dense.ai/share/session_20251227_090118_524452720570) · [PDF](/examples/session_20251227_090118_524452720570/writing_outputs/final/indian_heritage_tree_tourism_strategic_analysis.pdf)

**Scope.** Benchmarked California Redwoods against eight Indian heritage trees, nine analytic visuals, gap/branding/economic models, fifteen recommendations, then a thirty-three-page magazine-style strategic PDF with forty citations.

**Evidence.** ~33 pages, 13 figures. Datasets: redwood_stats.csv and indian_trees_stats.csv merged into comparative_matrix_normalized.csv with tourism scoring. Analyses: Normalized comparative scoring; revenue projection multipliers; strategic pillars with five-year $75M roadmap.

**Human estimate.** 200 hours × $175/hr = **$35,000** (5.0 full-time weeks). Breakdown: scoping/lit 28 h · data 24 h · analysis 32 h · figures 22 h · writing 82 h · review 12 h.

**K-Dense Web.** Mega tier, ~215 credits ≈ **$150** at Plus rate, ~6.0 hr wall clock.

**Savings.** $34,850 cost saved · 33× time compression · 233× cost multiplier.

_Notes: Thirteen PNG figures enumerated in writing_outputs/SUMMARY.md (magazine abstract, technical abstract, two AI schematics, fig01–fig09)._


### 34. India Crop Yield Prediction — Agriculture

**Links:** [Session](https://app.k-dense.ai/share/session_20251229_081931_9f2d364070e9) · [PDF](/examples/session_20251229_081931_9f2d364070e9/writing_outputs/final/indian_agriculture_analysis.pdf)

**Scope.** End-to-end ML on Indian crop and soil fertility datasets plus ANOVA and a revised two-column LaTeX paper with expanded citations and schematic figures.

**Evidence.** ~26 pages, 11 figures. Datasets: District crop production from GitHub (cleaned approx 237k rows) plus soil fertility CSV (100 samples); preprocessing, EDA, and RF regressor/classifier pipelines per manifest.. Analyses: RandomForestRegressor crop yield prediction (test R² about 0.71), RF soil fertility classifier, seasonal and regional ANOVA, correlation and geographic summaries..

**Human estimate.** 268 hours × $175/hr = **$46,900** (6.7 full-time weeks). Breakdown: scoping/lit 24 h · data 48 h · analysis 92 h · figures 32 h · writing 64 h · review 8 h.

**K-Dense Web.** Heavy tier, ~120 credits ≈ **$80** at Plus rate, ~3.0 hr wall clock.

**Savings.** $46,800 cost saved · 129× time compression · 597× cost multiplier.

_Notes: Page and figure totals match writing_outputs SUMMARY.md (v2: 26 pages, 11 figures). No PNG blobs are checked into this repo copy; counts follow session documentation and LaTeX references._


### 35. Solar Green Hydrogen Thermal Analysis — Energy

**Links:** [Session](https://app.k-dense.ai/share/session_20251230_095716_0b529465c6eb) · [PDF](/examples/session_20251230_095716_0b529465c6eb/writing_outputs/final/green_hydrogen_paper.pdf)

**Scope.** Hourly Rajasthan PV and electrolyzer simulation with degradation and net-impact waterfall narratives and a revised multi-column paper with dozens of techno-economic citations.

**Evidence.** ~19 pages, 13 figures. Datasets: 8760-row weather CSV, hourly PV DC power and electrolyzer CSVs plus degradation summaries and scenario comparison CSVs documented in manifest.json.. Analyses: pvlib PV simulation with thermal derating, PEM vs alkaline electrolyzer coupling, Arrhenius-style degradation coupling, Rajasthan vs baseline offset accounting..

**Human estimate.** 280 hours × $200/hr = **$56,000** (7.0 full-time weeks). Breakdown: scoping/lit 28 h · data 44 h · analysis 104 h · figures 40 h · writing 48 h · review 16 h.

**K-Dense Web.** Heavy tier, ~165 credits ≈ **$110** at Plus rate, ~3.2 hr wall clock.

**Savings.** $55,850 cost saved · 173× time compression · 527× cost multiplier.

_Notes: Uses SUMMARY.md v3 page count (19) and consolidated figures list spanning workflow schematics Sankeys and waterfall charts referenced across draft versions._


### 36. India EV Adoption Econometrics — Transportation

**Links:** [Session](https://app.k-dense.ai/share/session_20251230_101850_14dd57dfe39b) · [PDF](/examples/session_20251230_101850_14dd57dfe39b/writing_outputs/final/charging_ahead_ev_india.pdf)

**Scope.** Panel econometrics plus Monte Carlo TCO simulations written up as the 28-page Plug Beats Price report with substantive visualization and appendix tables.

**Evidence.** ~28 pages, 5 figures. Datasets: Synthetic 14-state 2019-2024 EV panel CSV (84 rows) documented in manifest plus simulation export files for tiered urban congestion scenarios.. Analyses: Panel RE vs FE diagnostics Hausman Monte Carlo TCO distributions infrastructure vs subsidy coefficient comparison cited in SUMMARY.md..

**Human estimate.** 272 hours × $200/hr = **$54,400** (6.8 full-time weeks). Breakdown: scoping/lit 24 h · data 44 h · analysis 96 h · figures 24 h · writing 72 h · review 12 h.

**K-Dense Web.** Heavy tier, ~160 credits ≈ **$105** at Plus rate, ~2.5 hr wall clock.

**Savings.** $54,250 cost saved · 181× time compression · 522× cost multiplier.

_Notes: Final page count pulled from SUMMARY.md (28-page v4). Figure count aligns with enumerated figures excluding backup graphical abstracts._


### 37. Carbon Removal Investment Pipeline — Climate Tech

**Links:** [Session](https://app.k-dense.ai/share/session_20251212_164706_1df6aaa34e0b)

**Scope.** Six-step Python diligence chain across grants patents publications network scraping talent validation selection then Markdown deck and abbreviated PDF pptx artifacts.

**Evidence.** ~26 pages, 0 figures. Datasets: NSF Grants OpenAlex/USPTO style JSON aggregates startup candidate CSV selections plus network metrics summarized in README.md outputs.. Analyses: Fuzzy cross-referencing momentum scoring collaborative graph stats TRL heuristics VC selection spreadsheets described across workflow scripts..

**Human estimate.** 476 hours × $225/hr = **$107,100** (11.9 full-time weeks). Breakdown: scoping/lit 48 h · data 100 h · analysis 180 h · figures 8 h · writing 120 h · review 20 h.

**K-Dense Web.** Mega tier, ~245 credits ≈ **$160** at Plus rate, ~2.0 hr wall clock.

**Savings.** $106,950 cost saved · 358× time compression · 670× cost multiplier.

_Notes: Repository PDF excerpt is three pages README notes while primary prose lives in sizeable Markdown memoranda so pdf_pages carries an estimated bundled memo plus slide-outline length for writing-hours parity with README-delivered corpus._


### 38. Infrastructure Impact on Women's Empowerment — Social Science

**Links:** [Session](https://app.k-dense.ai/share/session_20251229_112053_41c35bf38d4f) · [PDF](/examples/session_20251229_112053_41c35bf38d4f/writing_outputs/final/infrastructure_gender_research_v2.pdf)

**Scope.** India-wide policy-linked research blueprint integrating census NFHS PLFS UDISE RBI sources with causal designs equations and diagrams in a tightened 20-page v2 memo.

**Evidence.** ~20 pages, 7 figures. Datasets: Desk synthesis of Census 2011 NFHS PLFS UDISE RBI JJM references described rather than rebuilt panel microdata pulls per SUMMARY.md.. Analyses: Proposed IV 2SLS spatial DiD SEM pathways policy evaluation schematic reasoning without completed empirical estimation..

**Human estimate.** 284 hours × $175/hr = **$49,700** (7.1 full-time weeks). Breakdown: scoping/lit 40 h · data 56 h · analysis 100 h · figures 32 h · writing 48 h · review 8 h.

**K-Dense Web.** Heavy tier, ~135 credits ≈ **$90** at Plus rate, ~2.5 hr wall clock.

**Savings.** $49,550 cost saved · 227× time compression · 553× cost multiplier.

_Notes: Figures enumerated in SUMMARY correspond to schematic pathway diagrams SEM and DiD cartoons rather than empirical plots reflecting framework versus execution emphasis._


### 39. NGS Variant Callers in Clinical Practice — Genomics

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_173534_9bb5ad16d19b) · [PDF](/examples/session_20251224_173534_9bb5ad16d19b/writing_outputs/final/NGS_Variant_Callers_Review.pdf)

**Scope.** Twenty-three-page clinical genetics review consolidating germline somatic structural variant validation ACMG workflows and HIS integration citing four dozen bibliographic entries.

**Evidence.** ~23 pages, 8 figures. Datasets: None beyond curated literature references bibliography file supporting narrative benchmarking tables.. Analyses: Literature-derived comparisons of callers such as Mutect Strelka GATK DeepVariant CNV tooling plus QA validation guidance synthesis..

**Human estimate.** 296 hours × $175/hr = **$51,800** (7.4 full-time weeks). Breakdown: scoping/lit 56 h · data 24 h · analysis 120 h · figures 32 h · writing 56 h · review 8 h.

**K-Dense Web.** Heavy tier, ~150 credits ≈ **$100** at Plus rate, ~3.2 hr wall clock.

**Savings.** $51,650 cost saved · 274× time compression · 528× cost multiplier.

_Notes: Figure inventory matches SUMMARY.md eight pipeline diagrams aligning with graphical abstract plus numbered clinical workflow figures cited in LaTeX._


### 40. New Zealand Great Walks Planning — Travel

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_112534_0ab5bab9278d) · [PDF](/examples/session_20251224_112534_0ab5bab9278d/writing_outputs/final/nz_great_walks_trip_plan.pdf)

**Scope.** Sixty-five-day multimodal itinerary with budgeting training gear transport schematic figures embedded in twenty-three-page travel operations LaTeX report.

**Evidence.** ~23 pages, 6 figures. Datasets: Consolidated text notes on DOC logistics weather budgeting equipment drawn into narrative tables inside sources directory.. Analyses: Route timing optimization descriptive budgeting and phased training pacing rather than formal optimization solvers..

**Human estimate.** 220 hours × $125/hr = **$27,500** (5.5 full-time weeks). Breakdown: scoping/lit 28 h · data 56 h · analysis 48 h · figures 24 h · writing 48 h · review 16 h.

**K-Dense Web.** Heavy tier, ~125 credits ≈ **$85** at Plus rate, ~3.0 hr wall clock.

**Savings.** $27,400 cost saved · 147× time compression · 324× cost multiplier.

_Notes: SUMMARY cites six distinct graphics though LaTeX reuses timelines and budget visuals in appendix sections yielding duplicate includegraphics occurrences counted once operationally._


### 41. Flying Dutchman Historical Analysis — Maritime History

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_182332_0a76341055a3) · [PDF](/examples/session_20251224_182332_0a76341055a3/writing_outputs/final/flying_dutchman_paper.pdf)

**Scope.** Twenty-one-page interdisciplinary VOC maritime folklore paper linking archival voyage claims cultural remediation and meteorological explanations with six illustrative schematics.

**Evidence.** ~21 pages, 6 figures. Datasets: Archived research plaintext memos synthesized into bibliography-backed narrative.. Analyses: Historical argument plus navigation climate synthesis rather than computational modeling despite quantitative voyage comparisons referenced..

**Human estimate.** 280 hours × $150/hr = **$42,000** (7.0 full-time weeks). Breakdown: scoping/lit 48 h · data 40 h · analysis 100 h · figures 28 h · writing 56 h · review 8 h.

**K-Dense Web.** Heavy tier, ~140 credits ≈ **$95** at Plus rate, ~3.2 hr wall clock.

**Savings.** $41,850 cost saved · 172× time compression · 446× cost multiplier.


### 42. Great Pyramid Construction Analysis — Engineering

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_105449_4bd8fa278c83) · [PDF](/examples/session_20251224_105449_4bd8fa278c83/writing_outputs/final/pyramid_engineering_analysis.pdf)

**Scope.** Twenty-seven-page quantitative feasibility study contrasting ramp counterweight wet-sand and competing hypotheses with annotated engineering figures and bibliography.

**Evidence.** ~27 pages, 4 figures. Datasets: Literature-extracted dimensional energy and pacing assumptions enumerated as equations.. Analyses: Mass-throughput energetics reject linear ramp volumetric paradox arguments counterweight shear checks wet sand friction citations..

**Human estimate.** 296 hours × $200/hr = **$59,200** (7.4 full-time weeks). Breakdown: scoping/lit 32 h · data 40 h · analysis 120 h · figures 24 h · writing 68 h · review 12 h.

**K-Dense Web.** Heavy tier, ~145 credits ≈ **$95** at Plus rate, ~3.2 hr wall clock.

**Savings.** $59,100 cost saved · 274× time compression · 623× cost multiplier.


### 43. AI Predictions 2026 — AI/Technical

**Links:** [Session](https://app.k-dense.ai/share/session_20251231_185649_7c104f74b40a) · [PDF](/examples/session_20251231_185649_7c104f74b40a/writing_outputs/final/AI_Predictions_2026.pdf)

**Scope.** Twenty-four-slide executive deck summarizing sourced forecasts across agents silicon science regulation budgets with mirrored slide PNG plan documentation.

**Evidence.** ~24 pages, 24 figures. Datasets: Curated excerpts from Goldman Gartner Deloitte NVIDIA filings compiled in research summaries.. Analyses: Qualitative thematic synthesis benchmarking statistics without bespoke econometrics..

**Human estimate.** 312 hours × $200/hr = **$62,400** (7.8 full-time weeks). Breakdown: scoping/lit 88 h · data 40 h · analysis 64 h · figures 80 h · writing 24 h · review 16 h.

**K-Dense Web.** Mega tier, ~295 credits ≈ **$195** at Plus rate, ~40 min wall clock.

**Savings.** $62,150 cost saved · 744× time compression · 321× cost multiplier.

_Notes: Writing hours reflect one hour per SUMMARY-listed slide revision time captured under combined figure-slide polish totals per slide methodology mix._


### 44. Deadliest Animal Venoms Safety Guide — Biology/Safety

**Links:** [Session](https://app.k-dense.ai/share/session_20251224_181033_118bb9451f72) · [PDF](/examples/session_20251224_181033_118bb9451f72/writing_outputs/final/v1_venom_guide.pdf)

**Scope.** Sixteen-page field-oriented venom physiology and triage playbook covering five notorious species with bibliography but no raster figures checked into assets.

**Evidence.** ~16 pages, 0 figures. Datasets: None beyond cited clinical tox references.. Analyses: Structured mechanism reviews emergency stepwise protocols appendix tables distilled from guideline literature..

**Human estimate.** 216 hours × $175/hr = **$37,800** (5.4 full-time weeks). Breakdown: scoping/lit 40 h · data 32 h · analysis 92 h · figures 0 h · writing 44 h · review 8 h.

**K-Dense Web.** Medium tier, ~75 credits ≈ **$50** at Plus rate, ~90 min wall clock.

**Savings.** $37,700 cost saved · 240× time compression · 756× cost multiplier.

_Notes: SUMMARY.md acknowledges figures pending and LaTeX omits graphics so figure_count is zero pending packaging._


### 45. PDAC KRAS Mutation Survival Analysis — Cancer Biology

**Links:** [Session](https://app.k-dense.ai/share/session_20260116_102423_c4b0c6a77530) · [PDF](/examples/session_20260116_102423_c4b0c6a77530/writing_outputs/final/manuscript.pdf)

**Scope.** 17-page LaTeX manuscript on KRAS variants and survival in 2336 MSK PDAC patients with KM analysis and 23 verified refs

**Evidence.** ~17 pages, 5 figures. Datasets: cBioPortal PDAC MSK 2024 clinical and mutation TSVs TCGA supplemental cohorts. Analyses: KRAS frequency profiling log-rank KM survival cohort selection data aggregation.

**Human estimate.** 160 hours × $175/hr = **$28,000** (4.0 full-time weeks). Breakdown: scoping/lit 24 h · data 32 h · analysis 40 h · figures 12 h · writing 44 h · review 8 h.

**K-Dense Web.** Medium tier, ~65 credits ≈ **$50** at Plus rate, ~90 min wall clock.

**Savings.** $27,950 cost saved · 107× time compression · 560× cost multiplier.

_Notes: Figure count excludes the duplicate histogram PDF listed alongside PNGs in SUMMARY._


### 46. Biologically Inspired Robot Actuators PhD Proposal — Robotics

**Links:** [Session](https://app.k-dense.ai/share/session_20260119_113716_05b0fba1834d) · [PDF](/examples/session_20260119_113716_05b0fba1834d/writing_outputs/final/phd_research_proposal.pdf) · [Blog](/blog/ai-research-direction-discovery-phd-proposal)

**Scope.** 26-page PhD proposal with 35 plus citations seven figures covering hybrid actuation morphological computation and bio-mimicry

**Evidence.** ~26 pages, 7 figures. Datasets: Literature synthesis no primary experimental dataset. Analyses: Gap analysis three research tracks timeline and methodology framework.

**Human estimate.** 176 hours × $200/hr = **$35,200** (4.4 full-time weeks). Breakdown: scoping/lit 40 h · data 16 h · analysis 32 h · figures 12 h · writing 64 h · review 12 h.

**K-Dense Web.** Heavy tier, ~115 credits ≈ **$100** at Plus rate, ~45 min wall clock.

**Savings.** $35,100 cost saved · 235× time compression · 352× cost multiplier.

_Notes: Wall clock taken from blog post stating completion in about 45 minutes._


### 47. GBM Clinical Trial Landscape Analysis — Cancer Biology

**Links:** [Session](https://app.k-dense.ai/share/session_20260122_114521_a550ad5331dd) · [PDF](/examples/session_20260122_114521_a550ad5331dd/writing_outputs/final/GBM_Clinical_Trial_Landscape_Analysis.pdf) · [Blog](/blog/gbm-clinical-trial-landscape-analysis)

**Scope.** 21-page report on 1913 GBM interventional trials gap analysis sponsor landscape and seven figures

**Evidence.** ~21 pages, 7 figures. Datasets: ClinicalTrials.gov trial extract literature and blog companion docx. Analyses: Trial classification phase and modality distributions gap and attrition metrics.

**Human estimate.** 188 hours × $175/hr = **$32,900** (4.7 full-time weeks). Breakdown: scoping/lit 24 h · data 48 h · analysis 40 h · figures 16 h · writing 52 h · review 8 h.

**K-Dense Web.** Medium tier, ~70 credits ≈ **$50** at Plus rate, ~100 min wall clock.

**Savings.** $32,850 cost saved · 113× time compression · 658× cost multiplier.


### 48. Structure Therapeutics Investment Due Diligence — Biotech Investment

**Links:** [Session](https://app.k-dense.ai/share/session_20260127_141459_d78bbad6de57) · [PDF](/examples/session_20260127_141459_d78bbad6de57/writing_outputs/final/structure_therapeutics_due_diligence.pdf)

**Scope.** 52-page GPCR oral GLP-1 DD with NPV scenarios Porter SWOT patent clinical and safety dashboards sixteen figures

**Evidence.** ~52 pages, 16 figures. Datasets: Open Targets ClinicalTrials FAERS patents SEC filings preprints. Analyses: Risk-adjusted NPV competitive matrix safety signals IP runway market scenarios.

**Human estimate.** 404 hours × $300/hr = **$121,200** (10.1 full-time weeks). Breakdown: scoping/lit 32 h · data 72 h · analysis 120 h · figures 32 h · writing 128 h · review 20 h.

**K-Dense Web.** Mega tier, ~285 credits ≈ **$200** at Plus rate, ~6.0 hr wall clock.

**Savings.** $121,000 cost saved · 67× time compression · 606× cost multiplier.

_Notes: Figure count is the enumerated PNG list in SUMMARY sixteen distinct assets._


### 49. CoreWeave VC Due Diligence — Venture Capital

**Links:** [Session](https://app.k-dense.ai/share/session_20260127_143018_1e043a3c21a2) · [PDF](/examples/session_20260127_143018_1e043a3c21a2/writing_outputs/final/coreweave_due_diligence_report.pdf)

**Scope.** 63-page VC DD market sizing Porter unit economics NVIDIA supply chain risk and twelve scenario and sensitivity figures

**Evidence.** ~63 pages, 12 figures. Datasets: Market and competitor notes unit economics CSVs customer concentration risk matrix. Analyses: TAM funnel Porter Five Forces GPU unit model bull base bear customer concentration heatmaps.

**Human estimate.** 456 hours × $300/hr = **$136,800** (11.4 full-time weeks). Breakdown: scoping/lit 40 h · data 80 h · analysis 128 h · figures 24 h · writing 160 h · review 24 h.

**K-Dense Web.** Mega tier, ~250 credits ≈ **$150** at Plus rate, ~5.0 hr wall clock.

**Savings.** $136,650 cost saved · 91× time compression · 912× cost multiplier.

_Notes: Pages and twelve-figure gallery stated explicitly in SUMMARY._


### 50. Ramp Technologies VC Due Diligence — Venture Capital

**Links:** [Session](https://app.k-dense.ai/share/session_20260127_143328_800725940a1f) · [PDF](/examples/session_20260127_143328_800725940a1f/writing_outputs/final/ramp_due_diligence.pdf)

**Scope.** 44-page spend-management DD with TAM interchange economics retention risk heatmap and nine figures

**Evidence.** ~44 pages, 9 figures. Datasets: Market sizing competitive and regulatory text extracts interchange and retention metrics. Analyses: Unit economics waterfall LTV CAC framing regulatory risk matrix valuation trajectory.

**Human estimate.** 328 hours × $300/hr = **$98,400** (8.2 full-time weeks). Breakdown: scoping/lit 32 h · data 56 h · analysis 96 h · figures 16 h · writing 112 h · review 16 h.

**K-Dense Web.** Mega tier, ~220 credits ≈ **$150** at Plus rate, ~4.5 hr wall clock.

**Savings.** $98,250 cost saved · 73× time compression · 656× cost multiplier.

_Notes: SUMMARY reports forty four pages not sixty plus._


### 51. Viking Therapeutics (VK2735) Investment Analysis — Biotech Investment

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_085220_b017de51c5a6) · [PDF](/examples/session_20260203_085220_b017de51c5a6/writing_outputs/final/VKTX_Catalyst_Analysis_Report.pdf)

**Scope.** 27-page catalyst report Phase 3 VANQUISH competitive obesity benchmarks scenarios options overlays and eight figures

**Evidence.** ~27 pages, 8 figures. Datasets: Clinical trial registry clinical summaries patent and options CSVs FAERS references. Analyses: Probability-weighted valuation bull base bear safety matrix options payoff strategies.

**Human estimate.** 260 hours × $300/hr = **$78,000** (6.5 full-time weeks). Breakdown: scoping/lit 28 h · data 48 h · analysis 88 h · figures 16 h · writing 68 h · review 12 h.

**K-Dense Web.** Heavy tier, ~125 credits ≈ **$100** at Plus rate, ~2.3 hr wall clock.

**Savings.** $77,900 cost saved · 111× time compression · 780× cost multiplier.

_Notes: Eight figure filenames are listed though the section header says seven total._


### 52. Chegg (CHGG) Short Thesis Evaluation — Equity Research

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_085912_030e6e80a419) · [PDF](/examples/session_20260203_085912_030e6e80a419/writing_outputs/final/CHGG_Short_Thesis.pdf)

**Scope.** 15-page short memo AI disruption thesis financial deterioration DCF scenarios and eight charts

**Evidence.** ~15 pages, 8 figures. Datasets: CHGG filings market financials sentiment and technical indicator CSVs. Analyses: Revenue margin cash vs debt scenarios DCF sensitivity short interest and technicals.

**Human estimate.** 172 hours × $275/hr = **$47,300** (4.3 full-time weeks). Breakdown: scoping/lit 20 h · data 40 h · analysis 52 h · figures 12 h · writing 40 h · review 8 h.

**K-Dense Web.** Medium tier, ~50 credits ≈ **$50** at Plus rate, ~60 min wall clock.

**Savings.** $47,250 cost saved · 172× time compression · 946× cost multiplier.


### 53. Restaurant Industry Investment Analysis — Equity Research

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_085615_5f10d3d2357f) · [PDF](/examples/session_20260203_085615_5f10d3d2357f/writing_outputs/final/investment_memo.pdf)

**Scope.** 32-page long TOST short legacy POS pair-trade memo Rule-of-40 sensitivity eleven figures and twenty one refs

**Evidence.** ~32 pages, 11 figures. Datasets: FRED macro restaurant comps quarterly financials valuation multiples CSVs. Analyses: Share growth indexing margin scenarios EV multiples heatmap pair-trade thesis.

**Human estimate.** 236 hours × $275/hr = **$64,900** (5.9 full-time weeks). Breakdown: scoping/lit 24 h · data 48 h · analysis 56 h · figures 16 h · writing 80 h · review 12 h.

**K-Dense Web.** Heavy tier, ~145 credits ≈ **$100** at Plus rate, ~3.3 hr wall clock.

**Savings.** $64,800 cost saved · 71× time compression · 649× cost multiplier.


### 54. Cybersecurity M&A Target Screening — M&A Analysis

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_085757_d926b2c7d1fb) · [PDF](/examples/session_20260203_085757_d926b2c7d1fb/writing_outputs/final/cybersecurity_ma_screening_memo.pdf)

**Scope.** 14-page screening memo SentinelOne focus precedent multiples football field accretion strategic fit nine figures

**Evidence.** ~14 pages, 9 figures. Datasets: yfinance screen precedents accretion and valuation CSVs trade setup. Analyses: Target screen EV revenue precedents takeout premium accretion dilution acquirer scoring.

**Human estimate.** 184 hours × $325/hr = **$59,800** (4.6 full-time weeks). Breakdown: scoping/lit 24 h · data 32 h · analysis 64 h · figures 20 h · writing 36 h · review 8 h.

**K-Dense Web.** Medium tier, ~55 credits ≈ **$50** at Plus rate, ~75 min wall clock.

**Savings.** $59,750 cost saved · 147× time compression · 1.2K× cost multiplier.

_Notes: Nine PNGs counted as graphical abstract plus eight analytic figures documented in session README._


### 55. Sector Rotation in Fed Rate Cut Cycles — Macro Strategy

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_090059_a06c4b9b68c4) · [PDF](/examples/session_20260203_090059_a06c4b9b68c4/writing_outputs/final/macro_strategy_memo.pdf)

**Scope.** 31-page macro memo historical easing cycles sector rate sensitivity ETF flows trade ranking and fifteen figures

**Evidence.** ~31 pages, 15 figures. Datasets: FRED macro Yahoo ETF performance correlation and stress test CSVs. Analyses: Cycle comparison style performance sector screening rate sensitivity stress tests trade ranks.

**Human estimate.** 260 hours × $250/hr = **$65,000** (6.5 full-time weeks). Breakdown: scoping/lit 28 h · data 40 h · analysis 72 h · figures 32 h · writing 76 h · review 12 h.

**K-Dense Web.** Mega tier, ~185 credits ≈ **$100** at Plus rate, ~5.0 hr wall clock.

**Savings.** $64,900 cost saved · 52× time compression · 650× cost multiplier.


### 56. Resmetirom FDA Approval Probability — Biotech Investment

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_090952_ba7183dc7776) · [PDF](/examples/session_20260203_090952_ba7183dc7776/writing_outputs/final/fda_catalyst_brief_resmetirom.pdf)

**Scope.** 19-page LaTeX FDA catalyst brief on approved resmetirom (MASH): clinical vs OCA/elafibranor, regulatory path, commercial scenarios, SWOT, 10 embedded figures

**Evidence.** ~19 pages, 10 figures. Datasets: MAESTRO-NASH and competitor trials, FDA/AdCom context, epidemiology, SEC filings per SUMMARY. Analyses: Head-to-head efficacy/safety tables, market funnel and revenue scenarios, approval driver scoring, competitive positioning.

**Human estimate.** 168 hours × $300/hr = **$50,400** (4.2 full-time weeks). Breakdown: scoping/lit 24 h · data 32 h · analysis 40 h · figures 15 h · writing 48 h · review 7 h.

**K-Dense Web.** Heavy tier, ~115 credits ≈ **$100** at Plus rate, ~3.3 hr wall clock.

**Savings.** $50,300 cost saved · 50× time compression · 504× cost multiplier.

_Notes: Figure PNGs are not present under public/examples for this session; figure count is from SUMMARY and manifest paths._


### 57. HER2 ADC Competitive Intelligence — Biotech Investment

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_091112_0888c802c3bf) · [PDF](/examples/session_20260203_091112_0888c802c3bf/writing_outputs/final/HER2_ADC_Competitive_Intelligence_Report.pdf)

**Scope.** 22-page competitive intelligence PDF: 521 HER2 ADC trials, Open Targets/ChEMBL/FAERS, bull/bear thesis, 9 figures per SUMMARY

**Evidence.** ~22 pages, 9 figures. Datasets: ClinicalTrials.gov (521 trials), Open Targets ERBB2, ChEMBL payloads, FAERS ILD comparison per manifest. Analyses: Pipeline synthesis, payload differentiation, safety benchmarking, radar competitive matrix.

**Human estimate.** 196 hours × $300/hr = **$58,800** (4.9 full-time weeks). Breakdown: scoping/lit 28 h · data 40 h · analysis 48 h · figures 16 h · writing 56 h · review 8 h.

**K-Dense Web.** Medium tier, ~75 credits ≈ **$50** at Plus rate, ~100 min wall clock.

**Savings.** $58,750 cost saved · 118× time compression · 1.2K× cost multiplier.

_Notes: No committed PNGs in repo; figure count from writing_outputs/SUMMARY.md._


### 58. GLP-1 Receptor Agonists Safety Analysis — Biotech Investment

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_091322_5981a70f834a) · [PDF](/examples/session_20260203_091322_5981a70f834a/writing_outputs/final/GLP1_Safety_Signal_Report.pdf)

**Scope.** 23-page FAERS/Open Targets integrated safety report with Bonferroni-corrected pairwise tests and regulatory/investment implications, 8 figures

**Evidence.** ~23 pages, 8 figures. Datasets: FAERS ~655–657 reports, Open Targets GLP1R/GIPR per SUMMARY and manifest. Analyses: Signal quantification, pairwise OR tests with correction, mechanism integration.

**Human estimate.** 168 hours × $300/hr = **$50,400** (4.2 full-time weeks). Breakdown: scoping/lit 20 h · data 28 h · analysis 40 h · figures 14 h · writing 57 h · review 9 h.

**K-Dense Web.** Medium tier, ~70 credits ≈ **$50** at Plus rate, ~75 min wall clock.

**Savings.** $50,350 cost saved · 134× time compression · 1.0K× cost multiplier.

_Notes: Committed PNGs absent in repo; figures listed in SUMMARY and manifest outputs._


### 59. GLP-1 Adverse Events FAERS Analysis — Biotech Investment

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_092044_b80cb683d2ee) · [PDF](/examples/session_20260203_092044_b80cb683d2ee/writing_outputs/final/GLP1_Safety_Analysis_Presentation.pdf)

**Scope.** 17-slide FAERS deck: 20,307 reports, disproportionality, AE heatmap and time-series acceleration (thyroid vs gastroparesis), 5 key figures in SUMMARY

**Evidence.** ~17 pages, 5 figures. Datasets: OpenFDA FAERS 2004–2025, five incretin drugs per manifest and SUMMARY. Analyses: ROR/PRR, Mann-Kendall on proportions, comparative fingerprints.

**Human estimate.** 128 hours × $300/hr = **$38,400** (3.2 full-time weeks). Breakdown: scoping/lit 16 h · data 36 h · analysis 44 h · figures 10 h · writing 17 h · review 3 h.

**K-Dense Web.** Medium tier, ~65 credits ≈ **$50** at Plus rate, ~70 min wall clock.

**Savings.** $38,350 cost saved · 110× time compression · 768× cost multiplier.

_Notes: Slide deck; writing hours use 1 hr/slide. SUMMARY states ~25 min total project time for automation; PNGs not in writing_outputs in repo._


### 60. ADC Oncology Pair Trade Analysis — Biotech Investment

**Links:** [Session](https://app.k-dense.ai/share/session_20260203_092344_47a31388326a) · [PDF](/examples/session_20260203_092344_47a31388326a/writing_outputs/final/adc_oncology_pair_trade_report.pdf)

**Scope.** 22-page pair-trade research report: target landscape, technology scorecard, ILD safety, pipeline catalysts, long/short framework, 7 figures per SUMMARY

**Evidence.** ~22 pages, 7 figures. Datasets: Peer literature, trial IDs, market intelligence sources listed in SUMMARY. Analyses: Target and technology comparison, safety synthesis, investment quadrant.

**Human estimate.** 180 hours × $300/hr = **$54,000** (4.5 full-time weeks). Breakdown: scoping/lit 28 h · data 32 h · analysis 44 h · figures 12 h · writing 56 h · review 8 h.

**K-Dense Web.** Medium tier, ~40 credits ≈ **$50** at Plus rate, ~25 min wall clock.

**Savings.** $53,950 cost saved · 432× time compression · 1.1K× cost multiplier.

_Notes: SUMMARY documents ~25 min wall-clock for the automated pipeline; no PNGs committed under this session path._


### 61. U.S. Chocolate Market Valentine's Report — Market Research

**Links:** [Session](https://app.k-dense.ai/share/session_20260212_152520_8a0952fc765d) · [PDF](/examples/session_20260212_152520_8a0952fc765d/writing_outputs/final/US_Chocolate_Market_Report_Valentines_2026_v2.pdf)

**Scope.** 22-page v2 market report: tariff/cocoa forecasts, premium and e-commerce trends, 12 Valentine-themed figures, 37 references per SUMMARY

**Evidence.** ~22 pages, 12 figures. Datasets: Industry and trade sources summarized in SUMMARY v2 changelog. Analyses: Causal framework, demographic and channel trends, tariff price impacts.

**Human estimate.** 160 hours × $175/hr = **$28,000** (4.0 full-time weeks). Breakdown: scoping/lit 24 h · data 32 h · analysis 24 h · figures 18 h · writing 56 h · review 8 h.

**K-Dense Web.** Medium tier, ~70 credits ≈ **$50** at Plus rate, ~90 min wall clock.

**Savings.** $27,950 cost saved · 107× time compression · 560× cost multiplier.

_Notes: No figure PNGs in public/examples session tree; counts from SUMMARY.md only._


### 62. CD20×CD3 Bispecific Antibody Biomarker Discovery — Cancer Biology

**Links:** [Session](https://app.k-dense.ai/share/session_20260403_101458_7f548077d599) · [PDF](/examples/session_20260403_101458_7f548077d599/writing_outputs/final/biomarker_whitepaper_cd20xcd3.pdf) · [Blog](/blog/dlbcl-biomarker-discovery-bispecific-antibodies)

**Scope.** 34-page white paper integrating cBioPortal, DepMap, Open Targets, trials, FAERS; CREBBP–EZH2 co-mutation signal; Phase III SAP; 5 figures in PDF/SUMMARY

**Evidence.** ~34 pages, 5 figures. Datasets: 1184 DLBCL samples, DepMap lines, FAERS bispecific reports, 130 PubMed records per manifest. Analyses: Fisher/FDR co-occurrence, CRISPR CD20 stratification, composite biomarker scoring, SAP JSON.

**Human estimate.** 244 hours × $175/hr = **$42,700** (6.1 full-time weeks). Breakdown: scoping/lit 32 h · data 48 h · analysis 56 h · figures 10 h · writing 85 h · review 13 h.

**K-Dense Web.** Heavy tier, ~130 credits ≈ **$100** at Plus rate, ~3.7 hr wall clock.

**Savings.** $42,600 cost saved · 67× time compression · 427× cost multiplier.

_Notes: Unique PNG count in figures/ plus deduped writing_outputs/figures/ excluding _v1 matches five production figures._


### 63. PDAC Neoantigen Landscape & iNeST Strategy — Cancer Biology

**Links:** [Session](https://app.k-dense.ai/share/session_20260403_102916_6cb2547db0b9) · [PDF](/examples/session_20260403_102916_6cb2547db0b9/writing_outputs/final/PDAC_Neoantigen_Landscape_iNeST_Presentation.pdf) · [Blog](/blog/pdac-neoantigen-landscape-inest-strategy)

**Scope.** 36-slide deck: TCGA TMB, oncoplot, KRAS structure/RSA, neoantigen sankey, TME network, trials table, literature synthesis; 9 data figures in SUMMARY

**Evidence.** ~36 pages, 9 figures. Datasets: TCGA PAAD/LUAD/SKCM, PDB structures, ClinicalTrials + PubMed + Open Targets per manifest. Analyses: TMB stats, structural RSA, Sankey pipeline, stromal target network, trial comparison.

**Human estimate.** 180 hours × $175/hr = **$31,500** (4.5 full-time weeks). Breakdown: scoping/lit 28 h · data 40 h · analysis 52 h · figures 17 h · writing 36 h · review 5 h.

**K-Dense Web.** Heavy tier, ~120 credits ≈ **$100** at Plus rate, ~4.0 hr wall clock.

**Savings.** $31,400 cost saved · 45× time compression · 315× cost multiplier.

_Notes: Writing uses slide rule (1 hr/slide)._


### 64. CT-388 Obesity Drug Competitive Intelligence — Biotech Investment

**Links:** [Session](https://app.k-dense.ai/share/session_20260403_103155_c18ae3efc2c1) · [PDF](/examples/session_20260403_103155_c18ae3efc2c1/writing_outputs/final/CT388_Competitive_Intelligence_Report_2026.pdf) · [Blog](/blog/ct388-competitive-intelligence-obesity)

**Scope.** 44-page report: pipeline bubble, ChEMBL/lit pharmacology, efficacy bars, FAERS time series, patent Gantt, Open Targets genetics, SWOT and Phase III program per SUMMARY

**Evidence.** ~44 pages, 6 figures. Datasets: ClinicalTrials pipeline CSV, efficacy curation, FAERS PRR and time series, genetic_evidence.json per manifest. Analyses: Cross-drug efficacy normalization, FAERS surveillance, IP timeline, strategic synthesis JSON.

**Human estimate.** 256 hours × $300/hr = **$76,800** (6.4 full-time weeks). Breakdown: scoping/lit 32 h · data 40 h · analysis 48 h · figures 10 h · writing 110 h · review 17 h.

**K-Dense Web.** Mega tier, ~210 credits ≈ **$150** at Plus rate, ~5.0 hr wall clock.

**Savings.** $76,650 cost saved · 51× time compression · 512× cost multiplier.

_Notes: Six non-_v1 unique PNGs under figures/ and writing_outputs/figures/; SUMMARY lists six embedded figures._


### 65. ARPA-H Health Innovation Policy Report — Public Policy

**Links:** [Session](https://app.k-dense.ai/share/session_20260427_173439_56c05e499816) · [PDF](/examples/session_20260427_173439_56c05e499816/writing_outputs/final/ARPA-H_Policy_Report.pdf) · [Blog](/blog/arpa-h-policy-report-analysis)

**Scope.** 42-page policy report with budget/portfolio analysis, comparative ARPA model, SWOT, 7 recommendations, 3 scenarios, 3 case studies, 71 references, 14 figures per SUMMARY

**Evidence.** ~42 pages, 14 figures. Datasets: ARPA-H budgets/programs, statutes, GAO/CRS/NASEM and peer literature per SUMMARY. Analyses: Theory of change, Sankey portfolio flow, international comparison, SWOT and scenario planning.

**Human estimate.** 308 hours × $200/hr = **$61,600** (7.7 full-time weeks). Breakdown: scoping/lit 60 h · data 40 h · analysis 60 h · figures 28 h · writing 105 h · review 16 h.

**K-Dense Web.** Mega tier, ~280 credits ≈ **$185** at Plus rate, ~4.0 hr wall clock.

**Savings.** $61,415 cost saved · 77× time compression · 333× cost multiplier.

_Notes: Figure count uses 14 non-_v1 PNG basenames in writing_outputs/figures/._


### 66. Appalachian Lithium Resource Techno-Economic Analysis — Climate Tech

**Links:** [Session](https://app.k-dense.ai/share/session_20260501_125918_14ec7ecb40b5) · [PDF](/examples/session_20260501_125918_14ec7ecb40b5/writing_outputs/final/Appalachian_Lithium_TEA_Report.pdf)

**Scope.** 31-page TEA: USGS-style occurrence set, PyMC Monte Carlo, 10-year ramp model, IRA 45X, supply-chain offset; SUMMARY cites 10 embedded figures in PDF with 11 counted PNGs including graphical abstract

**Evidence.** ~31 pages, 11 figures. Datasets: Synthetic 160-occurrence pegmatite table, TEA benchmark CSV, comparables, ramp workbook per manifest. Analyses: Bayesian NPV/IRR, aeon ramp forecast, binding-constraint MC, geospatial viability index.

**Human estimate.** 244 hours × $225/hr = **$54,900** (6.1 full-time weeks). Breakdown: scoping/lit 28 h · data 40 h · analysis 64 h · figures 22 h · writing 78 h · review 12 h.

**K-Dense Web.** Heavy tier, ~180 credits ≈ **$120** at Plus rate, ~4.5 hr wall clock.

**Savings.** $54,780 cost saved · 54× time compression · 458× cost multiplier.

_Notes: Eleven unique non-_v1 PNGs in figures/ union writing_outputs/figures/; county viability is HTML only._


### 67. Oil-Shock Macro Stress Test — Macro Strategy

**Links:** [Session](https://app.k-dense.ai/share/session_20260505_134929_2f742528f8de) · [PDF](/examples/session_20260505_134929_2f742528f8de/writing_outputs/final/oil_shock_macro_stress_test_2026Q2.pdf)

**Scope.** 18-page macro memo: historical shock propagation, VAR(1) CPI scenarios ($110–150 Brent), EM vulnerability composite, sector trades, non-economist infographic; 27 refs per SUMMARY

**Evidence.** ~18 pages, 6 figures. Datasets: FRED monthly panel 1970–2026 (677 rows) per manifest, Treasury cross-refs. Analyses: Programmatic shock ID, VAR Cholesky forecasting with MC bands, EM rank-blended vulnerability.

**Human estimate.** 148 hours × $250/hr = **$37,000** (3.7 full-time weeks). Breakdown: scoping/lit 20 h · data 24 h · analysis 40 h · figures 13 h · writing 45 h · review 7 h.

**K-Dense Web.** Medium tier, ~80 credits ≈ **$50** at Plus rate, ~2.2 hr wall clock.

**Savings.** $36,950 cost saved · 68× time compression · 740× cost multiplier.

_Notes: Six unique non-_v1 PNGs in figures/ and writing_outputs/figures/._