Building Bitrise’s context-aware, browser-integrated AI copilot that keeps engineers in control

Bitrise developed a powerful AI assistant to enhance user productivity by acting as a co-pilot capable of handling broad, open-ended questions, identifying bottlenecks, and suggesting workflow optimizations. Initially exploring frameworks like LangChain and CrewAI, they opted for Pydantic.AI for its model-agnostic agent implementation and type-safe LLM responses but eventually transitioned to Go due to their engineering team's expertise in the language, allowing for seamless integration with existing systems and better production readiness. The AI agent, wrapped in a FastAPI-based HTTP API, supports tool approvals for state-modifying operations and is integrated into the Bitrise website, enabling both server-side and client-side tool executions. This approach ensures secure and efficient operation while leveraging the existing authentication infrastructure, ultimately achieving a robust AI-driven solution that enhances user interaction by controlling browser tabs for improved task management.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.