About the role
About the Role
Join the platform engineering team at an AI product studio and help build the foundation for reliable, high-performing AI products. You will shape how data, streaming responses, and language model requests move through production systems, helping the platform stay dependable as usage grows.
What You'll Do
* Integrate AI products with external data sources and multiple systems, handling rich context and real-time data flows.
* Route streaming responses and language model requests across providers, using fallback and cost logic to maintain correctness under production load.
* Monitor latency distributions, costs, and error rates, then act proactively to prevent system degradation.
* Make architectural decisions that improve reliability and performance while keeping systems simple and maintainable.
* Build and improve instrumentation and observability to identify missing signals and investigate complex production issues.
What We're Looking For
* At least 3 years of production backend engineering experience, including 3 or more years working with TypeScript.
* Experience owning production platform foundations and improving reliability and performance under real-world load.
* Production experience with Docker and Kubernetes.
* Strong system design skills, including the ability to reason about performance, latency, and cost trade-offs.
* Solid knowledge of PostgreSQL, APIs, queues, and caching.
* Experience with instrumentation, observability, and root-cause analysis; familiarity with streaming responses or language model request routing is valuable.
* Comfort contributing to frontend code when needed.
Compensation & Benefits
Salary range: €60,000 to €90,000 annually. Visa sponsorship is not available.
Location
Hybrid role based in Munich, Germany.