Sekai
AI Agent Engineer (Coding Agent)
San Francisco · Full-time · Posted Apr 30, 2026
About the job
Own the agent systems that generate interactive apps, improving their autonomy, reliability, quality, and production performance.
What you'll do
- Design the agent harness and long-horizon autonomous workflows.
- Build evaluation, regression-testing, and quality systems; design model routing and optimize reliability, cost, and latency.
- Add tracing, metrics, alerts, and observability; research new approaches and ship them to production.
What they ask for
- Experience building agentic systems involving tool use, orchestration, retry/repair loops, and context management.
- Experience with distributed or real-time systems.
- Strong software engineering fundamentals, with depth in Python and/or TypeScript.
- Track record of shipping and operating production systems and owning problems end to end.
Nice to have
- Built a customer-facing product from 0 to 1.
- Built a coding agent, devtool, IDE assistant, code-generation pipeline, evaluation framework, MCP-style tool protocol, or trained/fine-tuned LLMs.
Tools
Python, TypeScript, LangGraph, PydanticAI, Google ADK, Pi-mono, Vercel AI SDK
AI focus
Build and operate autonomous coding-agent systems for app generation, including model strategy, evaluation, reliability, and production deployment.
Model-written summary · the original posting is authoritative