# agentsite-static-version=e9078415-6885-4e34-b178-d9de692172ab
# cognition.com

> AI-optimized mirror of cognition.com containing 50 pages totalling 62,841 words of clean markdown content, structured data, and semantic HTML. Original source: https://cognition.com. Last updated: 2026-09-16T04:27:41.037Z.

Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG).

## Homepage

- [Cognition](/content/site-root.html): Cognition operates Devin, the first autonomous software engineer. Devin plans, writes, tests, and ships production code inside your existing workflows. (366 words)

## Articles & Blog Posts

- [Blog | Cognition](/content/blog/index.html): News, research, and engineering deep dives from Cognition, the team behind Devin, the autonomous software engineer. (3,014 words)
- [Cognition and AWS team up to help ambitious teams ship more, faster | Cognition](/content/blog/aws-sca/index.html): Cognition and AWS have entered a multi-year Strategic Collaboration Agreement to help enterprises deploy autonomous engineers in production, move legacy… (696 words, 2026-09-15)
- [Introducing Fusion in Devin Desktop & CLI | Cognition](/content/blog/local-fusion/index.html): Fusion is the most efficient frontier harness for Fable and Astra, up to 39% more efficient compared to other model harnesses across major coding… (1,050 words, 2026-09-11)
- [Introducing SWE-2: Pushing the Pareto Frontier | Cognition](/content/blog/swe-2/index.html): Today we’re introducing SWE-2, our most advanced coding model yet. SWE-2 delivers highly competitive agentic coding performance across multiple effort… (411 words, 2026-09-10)
- [Factoring RSA-260 | Cognition](/content/blog/factoring-rsa-260/index.html): How Devin and a Cognition researcher built the world’s highest-performance GPU lattice siever, to make factoring numbers 10x cheaper than the previous… (9,641 words, 2026-09-09)
- [Do it all with Devin: Announcing our Series E | Cognition](/content/blog/series-e/index.html): Cognition has raised over $2B at a $48B valuation, led by Andreessen Horowitz and Accel, to build the future of software engineering. (492 words, 2026-09-08)
- [One Year of Building Together | Cognition](/content/blog/one-year-of-building-together/index.html): One year ago, Cognition and Windsurf came together over the craziest 72 hours of our lives. Here’s what we’ve built together since. (856 words, 2026-07-14)
- [Devin is Now FedRAMP High In-Process, Unlocking Autonomous AI Engineering for Federal Agencies | Cognition](/content/blog/devin-fedramp-high-in-process/index.html): Cognition’s entire platform is now FedRAMP Class D (High) In-Process and listed on the FedRAMP Marketplace, giving engineering teams working in federal… (579 words, 2026-07-13)
- [SWE-1.7: Frontier Intelligence at a Fraction of the Cost | Cognition](/content/blog/swe-1-7/index.html): Today, we’re launching SWE-1.7, the most capable model we’ve trained so far. It reaches frontier-level intelligence at a much lower cost, advancing the… (568 words, 2026-07-08)
- [Measuring the Trustworthiness of Open-Source-Derived Models | Cognition](/content/blog/measuring-open-source-model-trustworthiness/index.html): We built an evaluation suite to assess model trustworthiness. Our results indicate that models developed from open-source models can be trusted, provided… (2,387 words, 2026-07-08)
- [FrontierCode 1.1 | Cognition](/content/blog/frontier-code-1-1.html): FrontierCode 1.1 refines our methodology to distinguish legitimate internet use from unfair use. (1,306 words, 2026-07-07)
- [Devin Fusion | Cognition](/content/blog/devin-fusion/index.html): Introducing Devin Fusion: a hybrid-model harness that keeps frontier-level coding intelligence while cutting costs with sidekick agents and dynamic… (1,676 words, 2026-06-29)
- [Introducing FrontierCode | Cognition](/content/blog/frontier-code/index.html): Today’s coding benchmarks have established that models can write correct code, but the question we should really be asking is: can models actually write… (2,518 words, 2026-06-08)
- [AI should earn its keep: Introducing the AI Productivity Guarantee | Cognition](/content/blog/ai-guarantee/index.html): Introducing the AI Productivity Guarantee for enterprise customers. If Devin delivers less engineering value than you’re paying for, Cognition will fund… (602 words, 2026-06-04)
- [Estimating the Productivity of an Autonomous AI Software Engineer | Cognition](/content/blog/ai-productivity/index.html): Engineering leaders want to know how much value AI is actually providing. We built a system that measures the number of human-equivalent hours of Devin's… (1,292 words, 2026-06-04)
- [Introducing Devin Desktop | Cognition](/content/blog/introducing-devin-desktop/index.html): The next generation of Windsurf, built around Devin Cloud, the Agent Command Center, and a full IDE for when you need to jump into the code. (378 words, 2026-06-02)
- [Verifying Agentic Development at Scale | Cognition](/content/blog/testing-development/index.html): What we’ve learned building end-to-end testing capabilities in Devin’s virtual machine (1,774 words, 2026-05-29)
- [More Devins in More Places | Cognition](/content/blog/series-d/index.html): Cognition has raised over $1B at a $26B valuation, led by Lux Capital, General Catalyst, and 8VC. (514 words, 2026-05-27)
- [Introducing Auto-Triage | Cognition](/content/blog/auto-triage/index.html): Devin can monitor for bugs, alerts, and incidents. When something breaks, Devin responds immediately, investigates with your tools, connects related… (539 words, 2026-05-18)
- [Devin now supports Android emulators | Cognition](/content/blog/android-emulator/index.html): Devin can now spin up an Android Virtual Device (AVD), enabling autonomous development for Android applications. (194 words, 2026-05-13)
- [What We Learned Building Cloud Agents | Cognition](/content/blog/what-we-learned-building-cloud-agents/index.html): Building cloud agents is difficult. We share some learnings from building Devin. (1,237 words, 2026-04-23)
- [Multi-Agents: What's Actually Working | Cognition](/content/blog/multi-agents-working/index.html): 10 months ago I argued against building multi-agent systems. Today, a narrower class works, where agents contribute intelligence while writes stay… (2,703 words, 2026-04-22)
- [New self-serve plans for Devin | Cognition](/content/blog/new-self-serve-plans-for-devin/index.html): We’re retiring the old Core and Team plans and introducing a new lineup: Free, Pro, Max, Teams, and Enterprise. We’re also beginning to charge for… (994 words, 2026-04-14)
- [Introducing SWE-Check: 10x Faster Bug Detection | Cognition](/content/blog/swe-check-10x-faster/index.html): Smaller specialized models can match or beat frontier generalists on the tasks they're trained for. Working with Applied Compute, we RL-trained SWE-check… (2,487 words, 2026-04-14)
- [An Early Preview of SWE-1.6 and Research Update | Cognition](/content/blog/swe-1-6-preview/index.html): We are sharing an early preview of our ongoing SWE-1.6 training run. (1,703 words, 2026-03-01)
- [How Cognition Uses Devin to Build Devin | Cognition](/content/blog/how-cognition-uses-devin-to-build-devin/index.html): Devin is a cloud agent platform for engineering teams. You work with it like a teammate — give it tasks, review its PRs, and let it handle your backlog… (1,853 words, 2026-02-27)
- [Closing the Agent Loop: Devin Autofixes Review Comments | Cognition](/content/blog/closing-the-agent-loop-devin-autofixes-review-comments.html): We built a feature that massively increased our internal token spend on Devin. But our PRs are now much more free of bugs and we can't go back. (519 words, 2026-02-10)
- [Agent Trace: Capturing the Context Graph of Code | Cognition](/content/blog/agent-trace/index.html): We’re excited to join in Cursor, Cloudflare, Vercel, git-ai, OpenCode and others in support of [Agent Trace](https://agent-trace.dev/). As described in… (759 words, 2026-01-29)
- [Devin's 2025 Performance Review: Learnings From 18 Months of Agents At Work | Cognition](/content/blog/devin-annual-performance-review-2025/index.html): Eighteen months since launch, Devin’s gone from tackling small projects, to deeply embedding in engineering teams at thousands of companies, including… (1,156 words, 2025-11-14)
- [Windsurf Codemaps: Understand Code, Before You Vibe It | Cognition](/content/blog/codemaps/index.html): Codemaps is meant to offer a shared understanding of a system between humans and AI, enabling your AI to teach you about the code you are looking at… (1,228 words, 2025-11-04)
- [Introducing SWE-1.5: Our Fast Agent Model | Cognition](/content/blog/swe-1-5/index.html): Today we’re releasing SWE-1.5, the latest in our family of models optimized for software engineering. It is a frontier-size model with hundreds of… (1,388 words, 2025-10-29)
- [Introducing SWE-grep and SWE-grep-mini: RL for Multi-Turn, Fast Context Retrieval | Cognition](/content/blog/swe-grep/index.html): We trained SWE-grep and SWE-grep-mini, fast agentic models specialized in highly parallel context retrieval. They match the retrieval capabilities of… (2,251 words, 2025-10-16)
- [Funding, growth, and the next frontier of AI coding agents  | Cognition](/content/blog/funding-growth-and-the-next-frontier-of-ai-coding-agents.html): Cognition has raised over $400M at a $10.2B post-money valuation led by Founders Fund. Other existing investors are similarly doubling down including Lux… (612 words, 2025-09-08)
- [How Eight Sleep Uses Devin as a Data Analyst | Cognition](/content/blog/how-eight-sleep-uses-devin-as-a-data-analyst/index.html): Building the future of sleep requires us to make data-driven product, operations, and finance decisions. But as a small team, we faced a challenge every… (1,077 words, 2025-09-04)
- [Build Your Own AI Data Analyst | Cognition](/content/blog/ai-data-analyst/index.html): Our #analytics Slack channel looks completely different than it did a year ago. Like most companies, we used to struggle with analytics. Even seemingly… (53 words, 2025-08-28)
- [Devin's MCP Marketplace | Cognition](/content/blog/mcp-marketplace/index.html): You can now connect your favorite MCP servers to Devin. (227 words, 2025-07-22)
- [Cognition’s acquisition of Windsurf | Cognition](/content/blog/windsurf/index.html): Cognition has signed a definitive agreement to acquire Windsurf, the agentic IDE. (637 words, 2025-07-14)
- [Coding Agents 101: The Art of Actually Getting Things Done | Cognition](/content/blog/coding-agents-101-the-art-of-actually-getting-things-done.html): The year is 2025. Coding agents aren't magic, but they're about the closest thing we have. We've noticed some engineers, in particular at the… (53 words, 2025-06-27)
- [Blockdiff: How we built our own file format for VM disk snapshots | Cognition](/content/blog/blockdiff/index.html): How we built blockdiff, an open-source tool for rapid block-level diffs and snapshots of VM disks. (2,143 words, 2025-06-23)
- [Don’t Build Multi-Agents | Cognition](/content/blog/dont-build-multi-agents/index.html): Frameworks for LLM Agents have been surprisingly disappointing. I want to offer some principles for building agents based on our own trial & error, and… (1,759 words, 2025-06-12)
- [The DeepWiki MCP Server | Cognition](/content/blog/deepwiki-mcp-server/index.html): The DeepWiki MCP server is here. Get codebase context and answers about any public GitHub repository in seconds. (244 words, 2025-05-22)
- [Kevin-32B: Multi-Turn RL for Writing CUDA Kernels | Cognition](/content/blog/kevin-32b/index.html): We explore reinforcement learning in a multi-turn setting, using intermediate feedback from the environment, and masking model thoughts to avoid exploding… (1,171 words, 2025-05-06)
- [DeepWiki: AI docs for any repo | Cognition](/content/blog/deepwiki/index.html): Turn any public GitHub repo into up-to-date documentation you can talk to, for free. (157 words, 2025-05-05)
- [Devin 2.0 | Cognition](/content/blog/devin-2/index.html): Introducing an agent-native IDE experience and a new Core plan starting at $20. (362 words, 2025-04-03)
- [Announcing The Devin Open Source Initiative | Cognition](/content/blog/cognition-open-source-initiative/index.html): OSS contributions Devin has already helped to make, and more about our Open Source Initiative. (443 words, 2024-12-12)
- [A review of OpenAI’s o1 and how we evaluate coding agents | Cognition](/content/blog/evaluating-coding-agents/index.html): Testing OpenAI’s newest o1 series of models with Devin for autonomous coding tasks (2,135 words, 2024-09-12)
- [SWE-bench technical report | Cognition](/content/blog/swe-bench-technical-report/index.html): One of our goals at Cognition is to enable Devin, an AI agent specializing in software development, to contribute code successfully to large, complex… (1,621 words, 2024-03-15)
- [Introducing Devin, the first AI software engineer | Cognition](/content/blog/introducing-devin/index.html): Setting a new state of the art on the SWE-bench coding benchmark. Meet Devin, the world’s first fully autonomous AI software engineer. (657 words, 2024-03-12)

## About Pages

- [About | Cognition](/content/about/index.html): Learn about Cognition, the applied AI lab behind Devin, and our mission to expand human capacity through autonomous software engineering. (359 words)

## Resources

- [Full Page Index](/index.html): Browse all cached pages with rich metadata
- [About This Cache](/about.html): Methodology, technical details, and usage guidelines
- [XML Sitemap](/sitemap.xml): Machine-readable sitemap for crawler discovery
- [Robots.txt](/robots.txt): Crawler directives
- [AgentSite Network](https://agentsite.network/network.html): Public index of AgentSites and their machine-readable resources