What they'll do
- Build and ship LLM-powered agents and workflows on the paved path (templates, pipeline, guardrails)
- Write and run evals for everything built; no system ships without passing its suite
- Integrate with GoodRx systems and data sources; instrument for cost and quality telemetry
- Pair with business users to iterate quickly inside bounded scopes
Must have
- 8+ years software engineering; 2.5+ building LLM applications in production (agents, RAG, tool use)
- Strong Python/TypeScript; API craftsmanship; prompt and context engineering that survives contact with users
- Can write LLM evals as part of building — non-negotiable
- Ships fast without skipping guardrails
Nice to have
- Retrieval/IR depth (for assistant-facing work); healthcare data handling; cost-optimization experience