Staff AI Platform Engineer
HybridSan Francisco, California, United States
San Francisco, California, United StatesHybridFull TimeSenior LevelHealthcare TechSmall
Full TimeSenior LevelSmallHealthcare Tech
Job Summary
Design and operate the knowledge graph and ontology that capture how grocery data relates, building retrieval systems using vector, graph, and structured stores to feed reliable context to models. Construct LLM-powered agents with tool-use and multi-step reasoning, serving them via infrastructure that runs reliably and cost-effectively. Stand up evaluation sets, LLM-as-judge harnesses, and observability pipelines to measure faithfulness, accuracy, and latency, then harden reusable capabilities from prototype to production using Databricks and MLflow.
Required Qualifications
- 5+ years building production software, data, or ML systems
- an excellent engineer with strong systems and API design (Python)
- Hands-on production experience with LLM systems: retrieval/RAG, agents and tool-use, prompt and context engineering — and, critically, evaluation
- Solid data-engineering and data-platform foundations: pipelines, data modeling, and a modern cloud data stack (Databricks/Spark, MLflow, cloud warehouses)
- Comfort in the messy middle of AI systems — retrieval quality, latency and cost trade-offs, non-determinism — and the instinct to build the guardrails and evals that make them trustworthy
- A platform mindset: you build for leverage and clean interfaces, and you thrive in ambiguity in a fast-moving space
- This is a hybrid role based in the San Francisco office (2 days/week)
- This position is not eligible for company sponsorship
Desired Qualifications
- Knowledge graphs, ontologies, or semantic layers in production; graph databases
- Vector stores (pgvector, Pinecone, Weaviate, etc.) and hybrid search
- MCP or similar tool/context protocols; agent frameworks (e.g., LangGraph)
- MLOps and model serving at scale; experimentation and observability tooling for LLM systems
- Experience in grocery, retail, or other complex enterprise data domains
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.