Senior Software Engineer (SDK)
at Langfuse
π Switzerland
π Germany
π Spain
π France
π United Kingdom
π Netherlands
π Zurich, Switzerland
π Munich, Germany
π Berlin, Germany
π Paris, France
π London, United Kingdom
π San Francisco, United States
π Germany
π Spain
π France
π United Kingdom
π Netherlands
π Zurich, Switzerland
π Munich, Germany
π Berlin, Germany
π Paris, France
π London, United Kingdom
π San Francisco, United States
EUR 90,000-160,000 per year
Used Tools & Technologies
Machine LearningRequired Skills & Competences
Tag name is followed by "@" symbol and proficiency level value.
About proficiency levels:
- 1-2 β basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 β daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 β you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 β exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
Go @ 4
Redis @ 4
TypeScript @ 4
Python @ 4
Java @ 4
Hiring @ 4
PostgreSQL @ 4
Next.js @ 4
API @ 4
LLM @ 4
OpenTelemetry @ 4
Observability @ 4
AI @ 4
Profiling @ 7
ClickHouse @ 4
GenAI @ 4
LangChain @ 4
- 1-2 β basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 β daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 β you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 β exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
Details
Open Source LLM engineering platform (Langfuse, now part of ClickHouse) focused on tracing, evaluation, and prompt management. Small, engineering-heavy team with offices in Berlin and San Francisco and hiring across EU timezones. Role expects a hybrid arrangement with roughly one week per month in the Berlin office.
Responsibilities
- Own and maintain Langfuse SDKs (Python and TypeScript) built on OpenTelemetry: reliability, performance, developer experience, error handling, batching, async flushing, and graceful degradation.
- Design APIs and developer-facing interfaces: naming, parameter structure, versioning, backwards compatibility, migration paths, and documentation.
- Integrate and maintain integrations with major AI frameworks and model providers (40+): OpenAI SDK, Vercel AI SDK, LangChain, LlamaIndex, Pydantic AI, OpenAI Agents, Amazon Bedrock AgentCore, LiveKit, and others; add new integrations as the ecosystem evolves.
- Maintain the OpenTelemetry (OTLP) endpoint path to support languages beyond Python/TypeScript (Java, Go, etc.) and collaborate with the OpenTelemetry community on GenAI conventions.
- Write and own public SDK documentation, integration guides, and migration paths; ship docs alongside SDK releases.
Requirements
- Strong software engineer excited about developer experience, API design, and building code that runs reliably in customers' production systems.
- Experience building or maintaining SDKs, client libraries, or developer tools, ideally in Python and/or TypeScript.
- Deep understanding of performance: profiling, benchmarking, and minimizing CPU/memory overhead in hot paths.
- Ability to manage projects independently: identify what to build, develop conviction, and ship features (RFCs, ownership of area, manage own Linear tasks).
- Thoughtful about versioning, backwards compatibility, and migration paths.
- Interest in open source and direct engagement with developers about integration challenges.
- Comfortable working in a small, accountable team where output is visible and matters.
- CS or quantitative degree preferred.
Bonus
- Experience with OpenTelemetry internals or observability instrumentation.
- Former founder or early startup experience.
- Contributions to popular open source SDKs or developer tools.
- ML/AI background or familiarity with the LLM framework ecosystem.
Tech Stack
- TypeScript monorepo (frontend: Next.js; backend/background: Express workers)
- PostgreSQL (transactional data), ClickHouse (tracing at scale), S3 (file storage), Redis (queues and caching)
- SDKs built on OpenTelemetry; support for many LLM frameworks and model providers
How We Ship / Process
- Ownership model: you identify what to build, propose solutions (RFCs), and ship them.
- Code reviews used as mentorship; maker schedule and focused asynchronous work.
- Hiring process can complete to offer in less than 7 days.
Compensation
- Salary range provided by the posting: 2220;90K 2220;160K (offers equity).