Job Description & Details
This is a heavy-hitting, 12-month contract role for an experienced AI/ML developer who knows how to ship production-grade Python code, not just Jupyter notebooks. You'll be based locally in St. Louis for an upfront F2F interview, building out serious generative AI and LLM infrastructure that actually works under load.
What You'll Actually Be Doing
You will spend your days architecting and building robust Retrieval-Augmented Generation (RAG) pipelines, integrating OpenAI APIs, and managing complex asynchronous workflows using FastAPI and asyncio. Your primary focus will be designing resilient backend systems that handle streaming data, tool calling, and structured JSON outputs reliably. You'll also deal with the unglamorous but vital side of software engineering: implementing circuit breakers, retries, and failure recovery so your AI applications don't fall over in production.
The Core Tech Stack
To succeed here, you need absolute mastery of Python in production environments, paired with at least a couple of years of hands-on experience building apps with LangChain, LangGraph, and modern LLM frameworks. You won't just be throwing prompts around; you need a solid grasp of embeddings, vector search, and re-ranking mechanisms. On the backend, heavy fluency in asynchronous Python libraries like FastAPI and httpx, alongside PostgreSQL paired with async SQLAlchemy and Alembic, is non-negotiable for managing state and data persistence.
Interview Expectations
Expect the hiring team to grill you on handling rate limits, latency, and partial failures when dealing with third-party LLM APIs. They will likely ask you to explain how you would design a fault-tolerant RAG pipeline that implements circuit breakers and fallback strategies when OpenAI goes down or throttles requests. They aren't looking for textbook definitions; they want to hear war stories about how you diagnosed race conditions in asynchronous Python code or optimized token usage and latency in a high-throughput multi-agent system.
Application Advice
When updating your resume for this role, skip the fluff about loving AI and focus entirely on your production battle scars. Make sure keywords like LangChain, LangGraph, FastAPI, asynchronous Python, and RAG pipelines are front and center, accompanied by metrics showing how you scaled previous systems. If you've built resilient microservices or managed async database migrations with SQLAlchemy in past contracts, highlight those experiences explicitly to clear the automated filters.