About this role
About the Role
Build reliable, production-ready agent capabilities that help users delegate complex work with confidence. This role combines agent systems engineering, platform development, integrations, and full-stack product work.
What You'll Do
- Evaluate advances in reasoning, planning, memory, tool use, and agent collaboration to identify useful product opportunities.
- Turn promising model behaviors into reusable skills and workflows that solve user problems.
- Build and improve Python agent infrastructure, including execution loops, orchestration, context management, structured outputs, retries, and recovery.
- Design tools and integrations for email, calendars, messaging, browsers, documents, CRMs, and business software.
- Own capabilities end to end across Python and Django services, APIs, React interfaces, data models, background jobs, observability, and operations.
- Develop reusable platform abstractions and improve execution latency, cost, reliability, and safety.
- Use evaluations, production traces, and user feedback to diagnose issues and improve agent behavior.
- Set technical direction for ambiguous problems and contribute to design reviews and engineering quality.
What We're Looking For
- At least 5 years of experience building, deploying, and operating production software systems end to end.
- Strong Python skills, including maintainable production code and experience debugging asynchronous systems.
- Hands-on experience building production agent systems, with knowledge of planning, tool calling, structured outputs, context and state management, retries, and orchestration.
- Full-stack experience with Django, APIs, React, and TypeScript.
- Experience building integrations with external APIs and business software platforms.
- Ability to turn evolving technical capabilities and loosely defined user needs into practical product features.
- Experience designing reusable platform abstractions and analyzing production traces across models, application code, databases, and APIs.
- Background working with large language models, prompt engineering, or model evaluation in production; experience optimizing distributed systems for latency, cost, and reliability is helpful.
- Strong computer science and engineering fundamentals, clear communication, and a track record of ownership and career progression or startup experience.
Location
On-site in Palo Alto, California, with hybrid flexibility. Visa sponsorship is not available.