← All roles
EngineeringSan Francisco · On-siteFull-time · On-siteSenior
AI / ML Full-Stack Engineer
Own the AI/ML layer end-to-end — model routing, retrieval, evals, fine-tuning — and ship it through the full Next.js + FastAPI stack into the product surface.
What you'll do
- Improve the multi-provider AI router (Groq / Anthropic / OpenAI / local)
- Own RAG pipelines, embeddings, vector storage, and retrieval quality
- Build evals and dashboards that measure quality across 30 product verticals
- Wire AI capabilities through API routes and into UI without hand-offs
- Drive measurable weekly improvements to latency, cost, and answer quality
What you bring
- 5+ years engineering with at least 2 in production ML/AI
- Strong Python plus working TypeScript / Next.js to ship UX yourself
- Hands-on with modern LLM APIs, vector DBs, and retrieval systems
- Reasons fluently about cost, latency, and quality trade-offs
- Comfortable owning a feature from inference to interface
Nice to have
- Built fine-tuning or distillation pipelines in production
- Experience with bandits, rerankers, or learning-to-rank
- Familiar with Cloud Run, Cloud SQL, and observability stacks
Compensation
Competitive salary plus meaningful equity. Calibrated on first call.
Apply
Send a short note — what you'd build, and a link to your best work. No cover letters.
Not the right role? See all openings.