Role: Senior AI Full Stack Engineer (Agentic Workflows & RAG) Location: Minneapolis, MN Contract position
Key Responsibilities
Agentic Orchestration: Design and implement complex "chains" and "graphs" (using LangGraph, LangChain, or equivalent) that manage multi-turn reasoning and tool-calling.
System Architecture: Develop the "connective tissue" between LLMs, Vector Databases (e.g., Pinecone, Weaviate), and legacy relational databases.
Production UX: Build React-based interfaces that handle asynchronous streaming, partial state updates, and "Human-in-the-loop" (HITL) correction UI.
Observability & Evaluation: Implement "LLM-Ops" patterns tracing requests (LangSmith/Arize), monitoring for hallucinations, and managing prompt versioning as code.
Secure Integration: Architect secure API gateways that handle rate-limiting, token-cost management, and PII masking before data reaches an LLM.
Technical Requirements
5+ Years Professional Engineering: Deep proficiency in Python (FastAPI/Pydantic) and at least one other ecosystem (Java/Spring or .NET).
Modern Frontend: 3+ years of React/TypeScript, specifically handling complex state management (Redux/Zustand) and real-time data (WebSockets/SSE).
Orchestration Mastery: Proven experience building Agentic workflows you must be able to explain how to handle loops, memory persistence, and tool-error recovery.
AI Patterns: Practical experience with RAG (Retrieval Augmented Generation) beyond the tutorial level handling chunking strategies, embedding drift, and re-ranking.
DevOps/SDLC: Strong Git hygiene, CI/CD experience, and a "test-first" mentality for both traditional code and LLM prompts.
Never miss an opportunity! Create an alert based on the job you applied for.
