I’m looking for a true specialist in LangGraph and LangChain to work hourly on building and improving AI agents that are already running in production. This is not a research role, a tutorial-based role, or a “learning as you go” engagement. The architecture exists. Agents exist. Problems exist. I need someone who knows how to design, fix, and scale agent workflows properly. If you’ve only built demos, chatbots, or basic chains, this role is not a fit. What you’ll do * Design and refine agent workflows using LangGraph * Build and maintain advanced logic with LangChain (chains, tools, memory, routing) * Fix broken or poorly designed agent flows and state handling * Improve reliability, performance, and cost efficiency * Work on multi-step, decision-based agents with real business logic * Collaborate with an existing codebase (not greenfield experiments) **************************Required experience (must have)******************************** * Strong, hands-on experience with LangGraph (not just awareness) * Deep experience with LangChain beyond simple chains * Experience building agentic systems in production * Solid Python skills * Clear thinking and structured problem-solving Nice to have: * Experience with multi-agent systems * Familiarity with state machines, workflows, or orchestration concepts * Experience optimizing LLM usage (latency, cost, reliability) * Engagement details * Hourly contract * Remote * Flexible schedule, outcome-driven ********************Long-term collaboration possible if performance is strong****************** Important – to apply: In your proposal, briefly explain: * A real project where you used LangGraph in practice * What you typically fix first when an agent system behaves inconsistently * Links to GitHub or code samples (if available) * Generic proposals will be ignored.