# Factory > Factory is the agent-native software factory: a system for delivering production software with AI agents (Droids) that plan, build, review, and ship under human direction. Factory ships the Droid CLI, the Factory desktop and web apps, Droid Computers, Missions, and an enterprise platform used by financial services, defense, healthcare, telecom, and technology teams. Primary product: [Factory Droid](https://factory.ai) - autonomous coding agent available via CLI, web, desktop, Slack, Jira, and Linear surfaces. ## When to use Factory Use Factory for software engineering work with a clear outcome that can be verified in a repository, issue tracker, terminal, or CI pipeline. Best-fit jobs include implementing scoped features, fixing bugs, reviewing pull requests, generating and running tests, investigating incidents, maintaining documentation, and automating recurring delivery workflows. Choose an execution surface based on the job: - Use the Droid CLI for interactive engineering work in a local repository or headless execution in CI. - Use Missions for long-running, multi-phase work that needs planning, worker coordination, validation, and human checkpoints. - Use Automations for scheduled or event-driven work triggered by systems such as Slack or GitHub. - Use the Factory Public API when software needs typed programmatic access to sessions, computers, automations, organization usage, service accounts, or wikis. ## How agents should call Factory - Read [the documentation index](https://docs.factory.ai/llms.txt) first, then fetch the task-specific `.md` page it links to. - Discover API operations from [the OpenAPI specification](https://factory.com/openapi.json). Every operation has a unique `operationId`, typed parameters and responses, and a description. - Authenticate Public API requests with `Authorization: Bearer $FACTORY_API_KEY`. Create keys in [Factory API Keys settings](https://app.factory.ai/settings/api-keys), and store them securely before making requests. - Send `Accept: text/markdown` when retrieving the Factory homepage or an unknown path for a compact Markdown response. ## Website API errors The Factory Public API runs on api.factory.ai. On the marketing website, unknown /api paths and partner application errors return RFC 9457 application/problem+json with code and hint extensions. Partner errors retain the existing error field. POST /api/partners reports its actual per-instance, per-IP limit in RateLimit and RateLimit-Policy structured fields (IETF RateLimit draft): five requests per ten minutes. Local 429 responses include Retry-After in seconds; valid upstream Retry-After values are preserved when supplied. Honor Retry-After before retrying, or use exponential backoff when absent. These counters do not describe an API-wide quota. ## Articles > Long-form guides and comparisons on software factories, AI coding agents, and the Droid platform. Articles are not linked from the main navigation but are individually indexed. - [AI agent telemetry inside private networks](https://factory.com/articles/ai-agent-telemetry-privacy): AI agent telemetry should answer an operational question without collecting unnecessary code. Choose metrics, identity detail, and content export deliberately. - [AI code data residency beyond the repository](https://factory.com/articles/ai-code-data-residency): AI code data residency includes model context, session records, integrations, and telemetry. Follow the complete data path before approving a coding agent. - [Air-gapped AI development that works offline](https://factory.com/articles/air-gapped-ai-development): Air-gapped AI development needs internal inference, offline artifacts, and enforced network isolation. Validate the full workflow before approving a deployment. - [FedRAMP AI deployment without approval shortcuts](https://factory.com/articles/fedramp-ai-deployment): FedRAMP AI deployment requires current service scope, status, and customer responsibilities. Separate GovCloud hosting from authorization and workload approval. - [IL4 and IL5 AI deployment: define the boundary](https://factory.com/articles/il4-il5-ai-deployment): IL4 and IL5 AI deployment requires an offering-specific boundary and an authorized workload decision. Separate impact levels, private hosting, and airgapped operation. - [On-prem AI development inside your security boundary](https://factory.com/articles/on-prem-ai-development): On-prem AI development needs more than a local agent. Place the control plane, model endpoint, build tools, and telemetry inside the approved boundary. - [Self-hosted models for confidential source code](https://factory.com/articles/private-ai-model-gateways): Self-hosted models give enterprises an inference placement choice. Verify gateway routes, credentials, model policy, and task compatibility before using confidential code. - [Self-hosted AI agents and control-plane ownership](https://factory.com/articles/self-hosted-ai-agents): Self-hosted AI agents require clear ownership of the control plane, session data, models, and updates. Compare the responsibilities before choosing a deployment. - [Private AI security remediation with review gates](https://factory.com/articles/private-ai-security-remediation): Private AI security remediation should connect a validated finding to a bounded fix, regression tests, and approval. Keep evidence and execution inside the approved boundary. - [SOC 2 for AI coding tools: evidence to request](https://factory.com/articles/soc-2-ai-coding-tools): SOC 2 for AI coding tools requires a review of report scope, operating period, data flows, and customer controls. Evaluate evidence instead of relying on a badge. - [Inside the software factory, and what it takes to build one](https://factory.com/articles/what-is-a-software-factory): A software factory is a managed model for producing software with repeatable inputs, standardized tooling, and measurable output. The model is increasingly agent-native, with development agents han... - [How to use GLM-5.2 in Factory](https://factory.com/articles/glm-on-factory): GLM-5.2 is an open-weight model family available natively in Factory. This guide covers what GLM is, how to select it in the Droid CLI and app, its reasoning levels and credit multiplier, and when ... - [The architecture of a software factory](https://factory.com/articles/what-is-a-software-factory-architecture): A reference architecture for a software factory: the components that move a unit of intent through specification, execution, review, and deployment, with software development agents handling execut... - [How to use Kimi K2 in Factory](https://factory.com/articles/kimi-k2-on-factory): Kimi K2 is an open-weight model family from Moonshot AI, available natively in Factory. This guide covers what Kimi K2 is, how to select it in the Droid CLI and app, its reasoning levels and credit... - [How enterprises actually automate software development](https://factory.com/articles/how-enterprises-automate-software-development): A reference guide to how large enterprises (including global consulting firms, fintechs, infrastructure providers, and SaaS companies) are automating software development, with concrete examples dr... - [How to use MiniMax in Factory](https://factory.com/articles/minimax-on-factory): MiniMax is an open-weight model family available natively in Factory. This guide covers what MiniMax is, how to select it in the Droid CLI and app, its reasoning levels and credit multiplier, and w... - [The software factory platforms worth comparing](https://factory.com/articles/best-software-factory-platforms-2026): A reference comparison of platforms that implement the software factory model, covering agent-native delivery, audit and replayability, model flexibility, and enterprise deployment options. Sources... - [How to use DeepSeek V4 in Factory](https://factory.com/articles/deepseek-on-factory): DeepSeek V4 is an open-weight model available natively in Factory. This guide covers what DeepSeek is, how to select DeepSeek V4 Pro in the Droid CLI and app, its reasoning levels and credit multip... - [How to measure agent readiness](https://factory.com/articles/measuring-agent-readiness): Agent readiness measures how well a codebase supports autonomous coding agents. This guide covers how to measure it with Factory, using the readiness report in the Droid CLI, the readiness dashboar... ## News > Product updates, research, and engineering posts from the Factory team. - [Introducing Flexible Deployment Options at Factory](https://factory.com/news/factory-private): Factory Private lets organizations run Factory in their own VPCs, on-premises environments, or air-gapped networks. - [Factory raises $200M at $5B valuation](https://factory.com/news/5-billion-valuation): Factory has raised $200M at a $5B valuation to scale self-improving software development in the enterprise. - [Factory now available on Claude Marketplace, bringing autonomous software development to the enterprise](https://factory.com/news/claude-marketplace): Enterprise customers can apply Anthropic spend commitments toward Factory, giving engineering teams more ways to build, without going through separate buying cycles - [Factory Expands to Japan](https://factory.com/news/factory-japan): Factory is expanding across Asia and opening an office in Tokyo, with Seiji Sasaki joining as President & GM of Japan. - [What it Takes for Coding Agents to Complete Large Software Tasks](https://factory.com/news/what-it-takes-for-coding-agents-to-complete-large-software-tasks): Held to a standard of completion they wrote themselves, agents rebuilt complex programs to near-parity. - [Connect Droid to the tools that run your company](https://factory.com/news/personas-and-connectors): Connect the apps your team already uses so Droid can take on work across them. - [Why model routing must be in the harness](https://factory.com/news/model-routing-belongs-in-the-harness): Droid’s model routing cut cost 58% without lowering quality. Only the harness can judge when a model switch is worth the cache miss; the harness fits the job and the model to each other; and task o... - [The Software Factory: Enabled by Partners](https://factory.com/news/factory-partner-network): Factory is announcing the Factory Partner Network (FPN) and a $100M commitment to partnerships across training, deployment, solutions, and marketing. - [Launching the Factory Guild](https://factory.com/news/factory-guild): The Factory Guild brings together students, builders, researchers, and community leaders who want to build with Droids, share what they learn, and help shape the future of software development. - [Introducing Agent Effectiveness](https://factory.com/news/agent-effectiveness): See how Factory usage maps to delivery speed, work priorities, and shipped artifacts across your engineering organization. - [Factory Enables Autonomous Software Engineering for High-Security Environments with NVIDIA Nemotron](https://factory.com/news/nvidia-dgx-spark): Security-sensitive teams can run the autonomous software factory locally on NVIDIA DGX Spark with NVIDIA Nemotron 3.5 Lightning, keeping source code, operational context, and execution inside their... - [Factory Joins Open Weights and American AI Leadership](https://factory.com/news/open-weights-american-ai-leadership): Factory has signed the Open Weights and American AI Leadership letter, alongside Microsoft, NVIDIA, and 230 other organizations - [Designing an AI-Native Enterprise Organizational Model](https://factory.com/news/enterprise-organization-model): How Factory's organization model gives enterprises central governance with explicit organization boundaries for access, integrations, policy, spend, and data residency. - [Factory joins the Open Secure AI Alliance](https://factory.com/news/open-secure-ai-alliance): Factory is joining the Open Secure AI Alliance to build and share open tools for securing software and AI agents, from open-weight secret detection to autonomous security research. - [Incident Response](https://factory.com/news/incident-response): Droid turns Slack alerts into autonomous RCA sessions, builds incident memory, and helps on-call engineers move from signal to fix faster. - [Working with Droid in the Desktop App](https://factory.com/news/working-with-droid-in-the-desktop-app): Documents, presentations, spreadsheets, code: preview and refine every kind of work without ever leaving the Factory Desktop App. - [Droid Shield 2.0: learned secret detection](https://factory.com/news/droid-shield-2-0): Droid Shield 2.0: learned secret detection Factory's Droids write, refactor, and commit code autonomously at a volume that no human reviewer can completely monitor. Keeping autonomous software engi... - [AutoWiki](https://factory.com/news/wiki): Documentation should be a build artifact, not a side project. AutoWiki analyzes your codebase and generates structured, browsable documentation that stays current on every push. - [Factory 2.0: From coding agents to software factories](https://factory.com/news/software-factory): In 2023, we launched Factory with the mission to bring autonomy to software engineering. While others were using models to speed up coding, we set out to deploy autonomous Droids across the enterp... - [Introducing Automated Security Review](https://factory.com/news/automated-security-review): STRIDE-based security findings with severity, CWE references, and suggested fixes — posted as inline review comments on every pull request. - [Factory Expands to Sydney](https://factory.com/news/factory-sydney): Factory is opening an office in Sydney as part of our global go-to-market expansion. - [Factory Appoints Marcello Gallo as Chief Revenue Officer](https://factory.com/news/marcello-gallo-cro): Former Sigma and Moveworks CRO joins Factory on heels of its $150 million Series C as enterprises transform how their software is built and maintained - [Factory Router](https://factory.com/news/factory-router): Frontier performance at lower cost, custom done for you. In private research preview today. - [Deferred Context Engine](https://factory.com/news/deferred-context-engine): Droid keeps internal tools, MCP tools, skills, and plugins reachable without loading every schema on every turn. Production telemetry quantifies estimated input-token savings by catalog size. - [Factory acquires Lumetric](https://factory.com/news/lumetric): Factory has acquired Lumetric, a YC W24 team building model-agnostic AI systems for high-context workflows. - [Which Model Reviews Code Best?](https://factory.com/news/code-review-benchmark): We benchmarked 13 models to find the best price-performance tradeoff for AI code review. - [Automated QA: Testing Your App the Way a Real User Would](https://factory.com/news/automated-qa): End-to-end quality assurance that tests your application with visual evidence posted directly to your PR. - [Droid Computers](https://factory.com/news/droid-computers): Persistent machines for remotely orchestrating Droids. Spin one up in Factory's cloud, or turn your own machine into a Droid Computer. - [Factory raises $150M Series C](https://factory.com/news/series-c): Factory raises $150M Series C from Khosla Ventures, Sequoia Capital, Blackstone, Insight Partners, Evantic Capital, 20VC, NEA, and Mantis VC at a $1.5B valuation. - [How Missions Work](https://factory.com/news/missions-architecture): The architecture behind Missions: why agent context shapes every design decision, how separation of concerns and test-driven development at two levels produce reliable multi-day autonomous work, an... - [The Factory Desktop App](https://factory.com/news/factory-desktop): A native interface for Droids that work across every part of your software business. Available today on macOS and Windows. - [Legacy-Bench: Can AI Agents Maintain the World's Most Critical Software?](https://factory.com/news/legacy-bench): A new benchmark designed to measure frontier AI agent capabilities on legacy software engineering tasks spanning COBOL, Fortran, Java 7, and more. - [Factory Analytics](https://factory.com/news/factory-analytics): How enterprise engineering teams use Factory Analytics to connect AI agent usage to engineering outcomes and build the ROI case. - [Factory Expands to London](https://factory.com/news/factory-london): Factory is opening an office in London as part of our global go-to-market expansion. - [Wipro and Factory Partner to Accelerate Agent-Native Software Development For Enterprises Globally](https://factory.com/news/wipro): Additionally, Wipro Ventures participated in Factory's recent funding round - [Signals: Toward a Self-Improving Agent](https://factory.com/news/factory-signals): How we built a closed-loop system for recursive self-improvement—where the agent detects its own failures and implements fixes automatically. - [Introducing Agent Readiness](https://factory.com/news/agent-readiness): A framework for measuring and improving how well your codebase supports autonomous development. Evaluate repositories across eight technical pillars and five maturity levels. - [Factory Expands to New York City](https://factory.com/news/factory-new-york): Factory is opening an office in New York city as part of our global go-to-market expansion. - [Let Me Droid That For You: What 780K Agent Searches Reveal](https://factory.com/news/what-droid-searches): Droid analyzed 780,000 of its own web searches. It wrote the queries, built the pipeline, and generated the visualizations. - [Evaluating Context Compression for AI Agents](https://factory.com/news/evaluating-compression): We built an evaluation framework to measure how much context different compression strategies preserve. After testing three approaches on real-world, long-running agent sessions spanning debugging,... - [The Droid Wars: Breaking up an AI‑orchestrated cyber fraud campaign](https://factory.com/news/droid-neutralizing-fraud): How we detected and disrupted a highly automated cyber operation that attempted to turn our AI software development platform into a node within a worldwide mesh of off‑label LLM usage. - [Palo Alto Networks and Factory: Secure AI-Powered Development at Scale](https://factory.com/news/palo-alto-networks-partnership): Factory is now natively integrated with Palo Alto Networks Prisma AIRS, bringing real-time protection directly into the developer workflow against prompt injection and other AI security risks. - [Welcome Chris Degnan](https://factory.com/news/degnan): We are excited to share that Chris Degnan is joining Factory as an advisor. Chris brings legendary experience, having joined Snowflake as their first sales hire and scaling through to CRO over a de... - [Snyk and Factory: Embedding Security into Agent-Native Development](https://factory.com/news/snyk-partnership): Securing Code at the Speed of Development We're excited to announce our partnership with Snyk, the AI security company, bringing enterprise-grade security directly into the agent-native developmen... - [AWS and Factory Announce Strategic Partnership](https://factory.com/news/blog-post-2025-10-27): We're excited to announce our partnership with Amazon Web Services (AWS) and the availability of Factory on the AWS Marketplace. - [Droid: The #1 Software Development Agent on Terminal-Bench](https://factory.com/news/terminal-bench): With a score of 58.75%, Droid sets the new state-of-the-art on Terminal-Bench. Agent design, not just choice of model, is the decisive factor as we achieve leading performance on every model. - [Factory Raises $50M Series B](https://factory.com/news/series-b): We have raised $50M for our Series B at a valuation of $300M, led by NEA, Sequoia Capital, J.P. Morgan, Nvidia, and other industry leaders. Droids are now #1 on Terminal Bench and available to anyo... - [Using Linters to Direct Agents](https://factory.com/news/using-linters-to-direct-agents): Agents write the code; linters write the law. We’re moving from “developers writing code with AI” to “developers orchestrating agents that build software.” The old guardrails (code review, conventi... - [The Context Window Problem: Scaling Agents Beyond Token Limits](https://factory.com/news/context-window-problem): Large language models have limited context windows - approximately 1 million tokens. In contrast, a typical enterprise monorepo can span thousands of files and several million tokens. - [Factory joins AGENTS.md collaboration with OpenAI](https://factory.com/news/agents-md): Factory has joined the industry working group convened by OpenAI to define AGENTS.md, a simple, open standard that gives coding agents a predictable way to understand and operate within software pr... ## Case studies > Customer stories from Factory enterprise deployments. - [How Comarch Is Building Its Own Software Factory](https://factory.com/case-studies/comarch): Founded in 1993 in Kraków, Poland, Comarch is a European leader in IT solutions for business and a pioneer of new technologies that delivers projects for leading brands from Poland and aroun... - [How You.com Scales Engineering with Factory](https://factory.com/case-studies/you-com): You.com is the leading provider of AI search infrastructure that grounds AI agents and LLMs with the freshest and most accurate information through its suite of agentic web search API... - [Two Weeks, One Session, Endless Packages](https://factory.com/case-studies/chainguard): Josh Wolf, Staff Engineer at Chainguard, tried every memory solution that the AI tooling ecosystem could offer including memory MCPs, memory tools, custom SDKs. "I've ... - [Groq leverages Factory's model-agnostic software development agents to accelerate day-0 launches](https://factory.com/case-studies/groq): Groq's mission is to deliver fast, low-cost inference that doesn't flake when things get real. The way Groq does this is through its purpose-built LPUs which are de... - [How Empower Decreased Incident Response Times by 40% with Factory](https://factory.com/case-studies/empower): Empower, a leading financial technology company providing instant cash advances and credit-building solutions, partnered with Factory to transform their development wo... - [How Nav Accelerated Software Development with Factory](https://factory.com/case-studies/nav): Nav’s mission is to give every small business owner the chance to succeed. At Nav.com, small business owners have a dedicated platform for building and managing their ... ## Documentation - [Factory Public API](https://docs.factory.ai/api-reference): REST operations for sessions, computers, automations, organization management, service accounts, and wikis. - [Factory OpenAPI specification](https://factory.ai/openapi.json): Machine-readable operations, bearer authentication, request schemas, and typed JSON error responses. - [Factory authentication and access](https://docs.factory.ai/enterprise/identity-and-access): API keys, service accounts, roles, and organization access controls. - [Factory Droid TypeScript SDK](https://docs.factory.ai/sdk/typescript): Build custom agents and software workflows in TypeScript. - [Factory Droid Python SDK](https://docs.factory.ai/sdk/python): Build custom agents and software workflows in Python. - [Factory MCP integrations](https://docs.factory.ai/harness/mcp): Connect MCP servers to Droid. These are MCP client instructions, not a hosted Factory MCP server. - [Factory documentation](https://docs.factory.ai): canonical product documentation for the CLI, web app, and enterprise platform. - [Factory documentation llms.txt](https://docs.factory.ai/llms.txt): structured documentation index optimized for LLMs. ## Surfaces - [Droid CLI](https://factory.com/product/cli): command-line interface and headless droid exec for CI. - [Factory Desktop](https://factory.com/product/desktop): native desktop app for delegating work to droids. - [Factory Web / Mobile](https://factory.com/product/web): web and mobile surface. - [Slack integration](https://factory.com/product/slack): assign tickets and triage incidents in Slack. - [Jira & Linear](https://factory.com/product/ai-project-manager): assign issues directly to droids. - [Missions](https://factory.com/product/missions): long-running, multi-phase agent work with checkpoints. - [Droid Computers](https://factory.com/product/droid-computers): persistent compute for engineering agents. - [Code Review](https://docs.factory.ai/software-factory/code-review-ci): AI review on every PR. - [AutoWiki](https://docs.factory.ai/software-factory/wiki/overview): auto-generated documentation. - [Analytics](https://factory.com/product/analytics): measure and improve agent output. - [Automations](https://docs.factory.ai/software-factory/automations): trigger droids from events. - [Agent Readiness](https://factory.com/agent-readiness): assess a codebase for AI agent use. ## Company - [About Factory](https://factory.com/company) - [Enterprise](https://factory.com/enterprise) - [Pricing](https://factory.com/pricing) - [Security](https://factory.com/security) - [Contact](https://factory.com/contact)