Own one or more agent surfaces end-to-end, from LLM integration and tool design to backend orchestration, conversational UI, and evals
Contribute to the shared agent platform: tool abstractions, chat persistence, eval frameworks, RAG, observability, permissioning
Shape technical direction on agent architecture, framework choice, and model strategy across Owner
Make thoughtful tradeoffs on latency, cost, and reliability so our agents feel fast and trustworthy in production
Raise the bar for agent quality through rigorous evals, thoughtful iteration, and patterns other engineers adopt

3+ years of software engineering experience, with meaningful time shipping production LLM-powered or agentic features
Fluency with tool calling, prompt engineering, and agent loops — you have built systems where the LLM is a first-class component, not a bolt-on
Experience with eval frameworks and observability for non-deterministic systems (Braintrust, LangSmith, custom eval harnesses, or equivalent)
Full-stack comfort — you can move between backend orchestration, tool design, and conversational UI without being blocked by hand-offs
High degree of cross-functional collaboration — past colleagues in product, design, and business would speak highly of your teamwork
Strong sense of ownership and drive, excellent communication, and a proactive approach to identifying and solving problems
Nice to have: Experience with RAG or retrieval systems at production quality
Nice to have: Background in restaurant tech, commerce, or SMB operator tooling
Nice to have: Contributions to open source agent frameworks, evals, or LLM tooling
Nice to have: Experience growing and mentoring engineers on agentic AI patterns

Software Engineer, Full-Stack

Key skills