Senior Full-Stack AI Engineer (FastAPI & Next.js) – SaaS LLM Feature Expansion

We are seeking an experienced Full-Stack AI Engineer to enhance and scale the AI features of our established, live SaaS platform. This is strictly not a build-from-scratch project. You will be joining an active production environment with an existing architecture. We need a pragmatic engineer who can quickly audit a running codebase, ship high-impact LLM functionality, and integrate clean solutions without disrupting or rewriting stable systems. Tech Stack Backend: Python (FastAPI), PostgreSQL, Redis, REST APIs, WebSockets Frontend: TypeScript, Next.js, React AI & Retrieval: OpenAI / Anthropic Claude / Gemini, Vector DBs (Pinecone, Qdrant, pgvector, or Weaviate) Key Deliverables & Responsibilities LLM Workflows & Agents: Build robust tool/function-calling pipelines, agentic workflows, structured data outputs, and error/retry fallbacks. Retrieval & Context: Implement and tune RAG pipelines, chunking strategies, embeddings, and vector search. Full-Stack Integration: Connect FastAPI-based AI microservices seamlessly to our TypeScript/Next.js frontend. Production Optimization: Monitor, evaluate, and tune for API latency, token efficiency, and response quality. Code Quality: Deliver clean, modular, and maintainable code that integrates smoothly with current services. What We Are Looking For Proven track record deploying production AI/LLM features into live applications. Production fluency in Python (FastAPI) and TypeScript (Next.js/React). Deep hands-on experience with LLM APIs, prompt engineering, structured outputs, and RAG architectures. Experience handling production edge cases: latency optimization, caching, rate limits, and fallback strategies. Familiarity with frameworks like LangChain, LangGraph, or LlamaIndex is a strong bonus. To Apply, Please Provide: Links to 1–2 production AI projects you built or contributed to (live links or GitHub repositories preferred). A brief summary of your hands-on experience with FastAPI, TypeScript, and RAG/vector search. Your availability and preferred hourly or contract rate. Start your proposal with the word "INTEGRATE" so we know you read the scope in full.