AI Engineer

Role overview A remote, full-time role owning the AI agent loop in production at a venture studio shipping consumer AI products. The engineer ships agent tools and workflows, manages routing and fallbacks across dozens of models, and builds the evaluations that keep the system honest.

Responsibilities - Ship agent tools and the workflows that chain them into real user outcomes - Own routing, fallback logic, and latency across 50+ models, including mid-stream failover - Build and maintain evaluations that catch regressions before users encounter them - Take new models from release announcement to production the same day - Investigate failures across provider outages, rate limits, silent regressions, and corrupted memory

Requirements - Production experience running agents or tool-calling systems at scale, with concrete examples of what broke and how it was detected - Habit of thinking in failure modes rather than happy paths - Benchmark-first mindset, applied even to your own work - Strong TypeScript skills and ability to pick up additional languages in days rather than weeks

Nice to have - Experience with MCP or similar connector ecosystems in production - Background in inference optimization or model serving

Benefits and work setup - Fully remote, full-time position