Key Responsibilities
- Translate ambiguous customer problems into working AI POCs at speed, demonstrating value before scaling investment
- Develop AI features and services using LLM APIs (OpenAI, Anthropic, Google) and self-hosted open-weight models (Llama, Qwen, Mistral)
- Design and implement agentic workflows with frameworks like LangGraph, CrewAI, or AutoGen, including tool use, planning, and multi-step reasoning
- Build production AI services and APIs in Python with FastAPI, handling streaming responses, async processing, and graceful degradation
- Implement RAG pipelines with vector databases (Pinecone, Weaviate, Qdrant) and integrate external tools via tool-calling
- Model per-request costs, track token usage, and optimize unit economics for AI features
Requirements
- At least one AI feature personally shipped to production with operational ownership
- 2-4 years of software/AI engineering experience with end-to-end production ownership
- Strong Python proficiency, API development with FastAPI, and production-grade coding practices
- Hands-on experience with LLM APIs and at least one agent framework (LangGraph, CrewAI, AutoGen)
- Working knowledge of RAG, embeddings, and vector databases