The opportunity
The LangSmith team owns and builds LangChain's core platform for observability, evaluation, and production reliability of AI systems. From tracing and annotation to run rules, evaluations, and beyond, this team owns LangSmith end-to-end.
What you'll do
Build and scale critical systems: design and operate high-throughput, data-intensive ingestion and trace-query systems supporting LangSmith, built on SmithDB , our purpose-built database for agent observability. Build monitoring, alerting, and automated recovery so the pipeline stays resilient.
Set API, SDK, and CLI standards: define and enforce the standards, tooling, and CI that power SDK generation across Python, TypeScript, Go, and Java; keep our developer surfaces consistent, high-quality, and self-served across feature teams.
Own integrations: build new integrations and maintain existing ones so it's easy to use LangSmith with any AI framework, agent, or tool — keeping us framework-agnostic
Solve complex problems: debug performance bottlenecks, optimize database queries, and architect solutions for distributed-system challenges
Respond to incidents: participate in an on-call rotation focused on post-incident learning, automation, and prevention
Platform engineering: hands-on experience designing and running data-intensive systems at scale
What they're looking for
- Developer experience: a track record of building high-quality, widely-adopted CLIs, SDKs, or API standards that developers actually enjoy using
- Database expertise: production experience with OSS datastores (PostgreSQL, Redis)
- Backend languages: Strong backend software engineering skills with production-level experience in Go, Python, or TypeScript.
- Infrastructure expertise: solid knowledge of cloud object storage, Kubernetes, containerized infrastructure, and cloud platforms (GCP, AWS)
