SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 150,000 - 190,000 / annual
LangChain is hiring an Agent Reliability Engineer to own the health, cost, performance, and business impact of their GTM Agent—a production AI system that powers go-to-market workflows across Sales, Marketing, Customer Success, and Support.
You'll be responsible for monitoring production health across every graph, catching errors, slow runs, expensive runs, and silent failures before they impact reps. You'll triage incoming issues from Slack, tickets, and rep reports, fixing small issues directly and routing complex problems to the right owner. Weekly, you'll run the eval suite, investigate failures, and convert real production bugs into permanent regression tests.
A core part of the role is tracking cost and latency by model, graph, use case, and role, then recommending concrete changes to model choice, reasoning effort, and caching strategies. You'll also track usage and adoption per rep and per feature, owning the weekly health report the team runs on. You'll build the business metrics that demonstrate leadership what the agent is worth—from reply rates and meetings booked to hours reclaimed and ROI.
Because LangChain builds the platform it runs on, you'll operate the agent on LangSmith the way they tell customers to, turning that practice into the reference story enterprises ask for. You'll build LangChain's own monitoring and alerting on LangSmith and write the "how we run our own agent" story for customers.
You'll work across Python 3.11, FastAPI, LangGraph, DeepAgents, LangSmith, Supabase Postgres, BigQuery, Anthropic and OpenAI models, and Slack and Next.js surfaces.
Required: strong production Python and SQL, real experience running LLM applications (tracing, evals, prompt and cache mechanics), SRE or production operations instincts (percentiles, SLOs, signal vs. noise), healthy skepticism about metrics, clear writing, and high agency.
Nice to have: LangGraph or LangSmith experience, eval suite building, BigQuery or dbt, DevRel-adjacent writing, empathy for sales and GTM users.
About LangChain
AI / Data / Infrastructure — developer platform and framework for building LLM applications and agents.
