The RoleYou'll be a core maintainer of the CrewAI open-source framework: the Crews and Flows APIs that hundreds of thousands of developers build on. This is engineering done in public. Your design decisions, code reviews, and API choices are visible to a large community and ripple out to everyone building on the platform.
You'll balance velocity with the long-term stewardship a widely adopted open-source project demands, and you'll treat developer experience as a first-class product.
What You'll Do- Design, build, and maintain core framework capabilities: agent orchestration, agent harness, Flows, task execution, tools, memory, MCP, A2A, and the public APIs developers depend on.
- Own the contributor experience: triage and review community PRs and issues, mentor contributors, and keep the bar high without killing momentum.
- Make and document API and DX decisions with backward compatibility and migration paths in mind. Changes here affect a very large installed base.
- Uphold engineering quality in the open: clean, well-tested Python, strong typing, and green CI on every change.
- Write and improve documentation, examples, and guides that help developers succeed quickly.
- Partner with runtime and platform engineers to keep the OSS-to-enterprise boundary coherent, and reconcile versioning so framework changes do not break downstream enterprise consumers.
- Engage the community directly through GitHub Discussions, forums, Discord, and other channels, turning real usage and feedback into roadmap signal.
RequirementsWhat We're Looking For- Strong Python engineering background, with a track record of shipping and maintaining library or framework code, not just application code.
- Genuine care for API design and developer experience. You've felt the difference between an API that's pleasant and one that isn't, and you can articulate why.
- Experience contributing to or maintaining open-source projects: reviewing external PRs, managing issue backlogs, and communicating decisions in public.
- Familiarity with the modern Python toolchain: uv, pytest, type hints, pre-commit, and well-tested code.
- Working knowledge of LLMs and agent concepts, or the appetite and ability to come up to speed fast.
- Comfort with the pace, ambiguity, and public feedback loops of an early-stage, high-growth environment.
