LangChain developer job description

About the role
We are hiring a Mid-Level LangChain Engineer to build production LLM features for remote-first UK and US product teams. You will work on RAG flows, structured outputs, tool calling, and Python services that use LangChain v1.x, LangGraph where appropriate, and standard observability practices.

What you'll do
Implement LangChain pipelines for retrieval, summarization, extraction, and assistant workflows. Connect vector stores, model providers, internal APIs, and typed output schemas. Add tests, traces, and evaluation examples for prompt and retriever changes. Debug production issues involving bad context, provider errors, rate limits, and schema failures. Collaborate with backend, product, and data teams to ship maintainable features.

What you'll bring
Strong Python experience with FastAPI or similar service frameworks. Hands-on use of LangChain, LCEL/Runnables, retrievers, tool calling, and structured output. Comfort with Pydantic v2, async code, unit testing, and basic cloud deployment patterns. Good judgment about what should be handled by code versus prompts.

Nice to have
Experience with LangSmith, LangGraph, vector database operations, document ingestion, prompt evaluation datasets, or migrating older LangChain code to current APIs.

About the role
We are hiring a Senior LangChain Engineer to own LLM workflows from design through production. You will lead architecture for retrieval, agents, evaluation, safety controls, and observability while writing hands-on Python code. The role suits an engineer who can turn ambiguous product ideas into reliable LangChain and LangGraph systems.

What you'll do
Design and implement stateful workflows with LangGraph, production RAG pipelines, model routing, structured output contracts, and safe tool integrations. Define tracing, evaluation, and release practices for LLM changes. Review prompts, retrievers, and agent tools for security, reliability, and maintainability. Diagnose issues across providers, vector stores, application code, and user behavior. Mentor engineers on modern LangChain v1.x patterns and pragmatic trade-offs.

What you'll bring
Deep Python backend experience and strong production LangChain knowledge. Experience with LangGraph state, checkpoints, retries, human-in-the-loop paths, and tool side effects. Ability to design evaluation sets, debug traces, and reason about latency, cost, privacy, and failure recovery. Clear communication with product and engineering stakeholders.

Nice to have
Experience supporting multiple LLM providers, warehouse or CRM tools, regulated data workflows, prompt-injection defenses, or platform libraries used by several teams.

About the role
We are hiring a Lead LangChain Engineer to set technical direction for a portfolio of LLM applications. You will guide architecture, standards, and delivery across product teams while remaining close to implementation. This role focuses on building a reliable platform for LangChain v1.x, LangGraph workflows, retrieval, evaluation, and safe tool use.

What you'll do
Define reference architectures, coding standards, observability defaults, and evaluation gates for LangChain services. Lead design reviews for agentic workflows, multi-tenant retrieval, provider routing, and human-in-the-loop operations. Establish patterns for secrets, tracing redaction, data retention, rollbacks, and incident response. Coach senior and mid-level engineers, unblock complex debugging, and align technical choices with product goals. Drive migration plans from legacy chains or experimental prototypes to maintainable production systems.

What you'll bring
Extensive Python and distributed systems experience, with proven ownership of production LLM applications. Strong knowledge of LangChain, LangGraph, LangSmith or equivalent tracing, Pydantic v2, retrieval systems, and model provider differences. Ability to influence without over-centralizing, write clear technical guidance, and make pragmatic build-versus-adopt decisions.

Nice to have
Experience building internal developer platforms, compliance-sensitive AI workflows, large document-processing systems, or shared evaluation frameworks for multiple product teams.