# K-Dense Web

> K-Dense Web is an AI agent that autonomously executes complex tasks across science, engineering, healthcare, finance, and beyond. It takes users from question to insight, problem to solution, by running multi-step workflows, executing real code, and generating publication-ready outputs.

For the comprehensive machine-readable context, see: https://www.k-dense.ai/llms-full.txt

## Quick facts

- **Company**: K-Dense Inc. — Palo Alto, CA (380 Portage Ave, Palo Alto, CA 94306)
- **Website**: https://www.k-dense.ai
- **Product**: K-Dense Web (https://app.k-dense.ai)
- **Tagline**: Research. Analyze. Synthesize.
- **Contact**: contact@k-dense.ai
- **Founded**: 2024
- **Documented use cases**: 79+
- **Blog posts**: 50+

## What it does

K-Dense Web is a fully hosted AI platform that autonomously:

1. **Analyzes data**: CSV, Excel, FASTA, PDB, SDF, and 200+ scientific formats.
2. **Searches and retrieves context**: 250+ databases plus hundreds of thousands of on-demand tools.
3. **Runs machine learning**: automatic model selection, hyperparameter tuning, SHAP interpretability.
4. **Generates professional outputs**: papers, slides, figures, manuscript-quality documentation.
5. **Supports multiple domains**: science, engineering, healthcare, finance, market research, and more.

## Key differentiators

- End-to-end pipeline from raw data to publication.
- Executes real Python/R/ML code (not just generates text).
- Intelligent model selection and hyperparameter optimization.
- Grounded outputs with reduced hallucinations.
- 79+ documented real-world use cases.
- Both open-source and hosted offerings.

## Pricing overview

- **Open-source packages**: Free. See "Open source" section below and each repository for license details.
- **K-Dense Web Personal**: Pay-as-you-go.
- **K-Dense Web Plus**: $199/month (300 credits/month).
- **K-Dense Web Team**: $499/month (800 credits/month, unlimited seats).
- **K-Dense Enterprise**: Custom pricing, integrations, SOC 2, HIPAA-ready, SSO, on-prem options.
- **Research Grant Program**: 90% off Team plan for academic labs and non-profit researchers.

## Open source

K-Dense ships 10 selected open-source libraries, skills, and tools for scientific agents, research automation, writing, data science, molecular optimization, and machine learning:

- **scientific-agent-skills** — 141 ready-to-use scientific and research Agent Skills for biology, chemistry, medicine, scientific computing, and 78+ databases. → https://github.com/K-Dense-AI/scientific-agent-skills
- **claude-scientific-writer** — Deep research and scientific writing tool for publication-ready academic documents with literature lookup, citation verification, and CLI/plugin workflows. → https://github.com/K-Dense-AI/claude-scientific-writer
- **k-dense-byok** — Desktop AI research assistant powered by your own API keys, specialist agents, parallel chats, OpenRouter models, and local Ollama support. → https://github.com/K-Dense-AI/k-dense-byok
- **mimeo** — CLI that researches an expert's public work, clusters their mental models, and emits a SKILL.md or AGENTS.md for coding agents. → https://github.com/K-Dense-AI/mimeo
- **mimeographs** — 80 ready-to-use expert SKILL.md and AGENTS.md profiles for founders, philosophers, scientists, and AI researchers. → https://github.com/K-Dense-AI/mimeographs
- **scientific-agents** — Research-grounded AGENTS.md profiles that turn generic agents into domain-aware scientific and engineering practitioners. → https://github.com/K-Dense-AI/scientific-agents
- **science-superpowers** — Composable methodology skills that guide research agents through question framing, literature grounding, analysis design, and pre-registration. → https://github.com/K-Dense-AI/science-superpowers
- **rowan-autosearch** — Agent-driven molecular optimization harness for Rowan workflows, RDKit constraints, auditable candidate logs, and HTML reports. → https://github.com/K-Dense-AI/rowan-autosearch
- **karpathy** — Agentic machine learning engineer built with the Claude Agent SDK, Google ADK, and Scientific Agent Skills. → https://github.com/K-Dense-AI/karpathy
- **agentic-data-scientist** — Adaptive multi-agent data science framework with planning, execution, validation, reflection, MCP integration, and Scientific Agent Skills. → https://github.com/K-Dense-AI/agentic-data-scientist

All repositories live under the K-Dense GitHub organization: https://github.com/K-Dense-AI

## Common questions

- **What is K-Dense?** An AI agent platform for autonomous, end-to-end complex task execution across many domains.
- **Who is it for?** Researchers, analysts, engineers, and operators who need to turn data and questions into decisions and publication-ready outputs.
- **How does billing work?** Pay-as-you-go credits (non-expiring) or monthly/annual subscriptions with refreshing credits.
- **Is my data private?** Yes. Data is encrypted in transit and at rest; we do not train on user data. Enterprise offers private cloud and on-prem options.
- **What about open source?** K-Dense publishes selected open-source projects for Agent Skills-compatible tools, research agents, data science, machine learning, and scientific writing — see the "Open source" section above.

## Key pages

- [Homepage](https://www.k-dense.ai)
- [About](https://www.k-dense.ai/about): Company mission, team, values
- [Use Cases](https://www.k-dense.ai/use-cases): 79+ real research examples
- [Pricing](https://www.k-dense.ai/pricing): Personal, Plus, Team, Enterprise, and open-source
- [Enterprise](https://www.k-dense.ai/enterprise): Custom integrations & deployment
- [Research Grant Program](https://www.k-dense.ai/research-program): 90% off Team plan for research orgs
- [Blog](https://www.k-dense.ai/blog): Insights & updates
- [Tutorials](https://www.k-dense.ai/tutorials): Step-by-step video tutorials
- [Newsletter](https://www.k-dense.ai/newsletter): AI research updates
- [FAQ](https://www.k-dense.ai/faq): Frequently asked questions

## Recent blog posts

- [Benchmarking Google's Omni Flash for Scientific Video](https://www.k-dense.ai/blog/benchmarking-google-omni-flash-scientific-video) — A 35-case K-Dense benchmark of Google's Omni Flash shows stunning scientific video quality, but unreliable text, physics, and factual correctness.
- [Benchmarking Nano Banana 2 Lite for Scientific Image Generation](https://www.k-dense.ai/blog/benchmarking-nano-banana-2-lite-scientific-image-model) — A 240-image benchmark of four scientific image models shows Nano Banana 2 Lite is fastest, while GPT Image 2 still leads on raw figure quality.
- [AI Scientists Need Lab Escape Rooms, Not More Exams](https://www.k-dense.ai/blog/science-needs-better-black-boxes) — The next AI scientist benchmark should be a lab escape room: a fresh hidden world, limited probes, real evidence, and no place for science theater to hide.
- [Benchmarking NVIDIA BioNeMo Agent Toolkit Skills for NIM microservices](https://www.k-dense.ai/blog/benchmarking-nvidia-bionemo-nim-skill) — A controlled benchmark of 10 NVIDIA BioNeMo Agent Toolkit skills for NIM microservices shows where skills win: routing, cost, scale, and weak-model reliability.
- [Reproduction, Not Generation, Is AI's Killer App for Science](https://www.k-dense.ai/blog/reproduction-not-generation-ai-for-science) — AI is flooding science with claims no one can check. The quieter story: agents now reproduce most published findings — and that is the bigger deal.
- [Can an AI Agent Run Your Mass Spec Pipeline? Benchmarking the PyOpenMS Skill](https://www.k-dense.ai/blog/benchmarking-pyopenms-skill-mass-spectrometry) — A reproducible 250-run study of feature detection, adduct grouping, quantification, and identification, run by an AI agent with and without the pyOpenMS skill.
- [The Week Science Models Became Real](https://www.k-dense.ai/blog/frontier-science-models-arrive) — Fable 5 and GPT-Rosalind show frontier AI moving into scientific workflows. The next bottleneck is not intelligence, but evidence.
- [Beyond RDKit: Benchmarking the Rowan Agent Skill Against Experiment](https://www.k-dense.ai/blog/benchmarking-rowan-skill-chemistry) — A reproducible study of pKa, logD, tautomers, ADME, and docking run by an AI agent with the Rowan skill, measured against RDKit and experimental ground truth.

Full post text is available at `https://www.k-dense.ai/blog/<slug>.md` in raw markdown.

## Security & compliance

- Data is encrypted in transit and at rest.
- K-Dense does not train on user data.
- Enterprise supports SOC 2 Type II, HIPAA-ready infrastructure, SSO/SAML, private cloud, and on-premises deployment options.

## Social

- GitHub: https://github.com/K-Dense-AI
- Twitter/X: https://x.com/k_dense_ai
- LinkedIn: https://www.linkedin.com/company/k-dense-inc
- YouTube: https://www.youtube.com/@K-Dense-Inc

## Machine-readable resources

- Sitemap: https://www.k-dense.ai/sitemap.xml
- RSS feed: https://www.k-dense.ai/feed.xml
- Summary LLM context: https://www.k-dense.ai/llms.txt
- Well-known LLM context: https://www.k-dense.ai/.well-known/llms.txt
- Full LLM context: https://www.k-dense.ai/llms-full.txt
- Typo-safe LLM context alias: https://www.k-dense.ai/llm.txt
- Blog raw markdown: https://www.k-dense.ai/blog/<slug>.md

---
Last updated: 2026-06-30