Factory is an agent-native platform for software development, whose core is an autonomous software factory driven by agents (called Droids). Users can use these agents to plan, build, review, and ship software in the desktop app, terminal, web, and CLI. The platform is positioned to enable software to improve itself and to help teams build autonomous software factories.
Main capabilities include:
- Autonomous planning and building: Droids can plan, build, and verify changes, while teams set goals and policies.
- Full development lifecycle coverage: From planning, building, and review to delivery, it covers multiple stages of the software development lifecycle.
- Multi-entry access: Supports desktop app, terminal, web, CLI, as well as IDE, CI pipelines, and Slack.
- Governance and sovereignty control: Users can control where the system runs, how it learns, and which models it uses, enabling model-agnostic agents.
- Continuous self-improvement: Deploy governed systems that continuously turn enterprise signals into verified production software.
- Business value measurement: Connect engineering adoption and effectiveness to growth, profitability, risk, and business performance.
- Model routing: Automatically route each task to the most suitable model, balancing quality, speed, and cost.
- Code review: Thoroughly review every pull request, catching real issues before merge through inline comments.
- Automatic documentation: Provide every code repository with a wiki that updates in real time as the code changes (AutoWiki), making it easy for anyone to browse the system.
- Automation: Let repetitive engineering work run automatically, with Droids planning, executing, and verifying in a shared context.
- Analytics and monitoring: Track adoption and consumption by developer and team, provide real-time usage metrics and exportable reports; monitor active tasks, track work progress, and view detailed execution logs through Mission Control.
- Readiness assessment: Measure an organization's readiness score through Agent Readiness, and track changes for each code repository over time.
- Specialized hardware: Droid Computers are purpose-built hardware that run agents locally on dedicated compute, keeping state persistent across sessions.
How to use: Users can invoke Droids through entry points such as the desktop app, Droid CLI, IDE, CI pipelines, and Slack, letting them plan, execute, and verify tasks in a shared context. The platform can help teams turn enterprise intent and codebase context into verified production software.
Use cases: Suitable for engineering teams, especially large enterprises that want to standardize governed, multi-model agents across the entire organization.It can be used to automate coding, testing, and deployment, improve software delivery efficiency, and connect engineering metrics with business performance.
Limitations: The materials do not provide specific pricing, costs, platform support details, or usage steps; this information is not currently available.
