Description
We're building production AI applications and agentic systems, and we're looking for a full-stack engineer who is genuinely comfortable working across the AI stack — not just calling an LLM API, but designing reliable, stateful agent workflows. You'll work on real products where the AI layer needs to be robust, observable, and maintainable. What youll do : Design and build agentic workflows using LangGraph (stateful graphs, multi-agent orchestration, human-in-the-loop patterns) Develop LLM-powered features using LangChain and the Claude SDK (Anthropic's Python SDK) — tool calling, structured outputs, prompt engineering, and streaming Build and integrate RAG pipelines (chunking, embeddings, vector databases, retrieval optimization) Write clean, well-structured Python backend code (FastAPI or similar) and integrate it with a modern frontend Handle context management, token/cost optimization, guardrails, and error handling in production Collaborate on architecture decisions and ship features end to end Required skills : Strong Python fundamentals and backend development experience Hands-on experience with LangGraph — you can explain nodes, edges, state, and why you'd choose a graph over a simple chain Practical experience with LangChain Experience with the Claude SDK / Anthropic API (tool use, system prompts, structured responses) Understanding of RAG systems and vector databases (Pinecone, pgvector, Chroma, etc.) Comfortable with API design, async patterns, and clean code pract