LLM Platform Engineer for AI SaaS

The company is an AI SaaS and Developer Tools business based in Berlin. It develops LLM infrastructure that enables engineering teams to deploy, evaluate, observe, and operate LLM-powered applications at scale.

• Build platform services for production LLM applications;
• Develop model gateways and APIs for accessing multiple LLM providers;
• Build shared RAG and vector retrieval infrastructure;
• Develop Python tooling and services for AI engineering teams;
• Deploy and operate AI workloads on Kubernetes;
• Build evaluation pipelines for LLM and RAG applications;
• Implement observability across prompts, models, retrieval, latency, cost, and failures;
• Automate infrastructure and environments using Terraform;
• Create deployment tooling and self-service capabilities for engineering teams;
• Improve reliability, scalability, and operational standards across production AI applications.

• 4+ Years in Platform Engineering, AI Engineering, MLOps, Backend Engineering, or similar roles;
• Strong understanding of production cloud-native systems;
• Nice to have: OpenAI, Anthropic, or multiple model providers, Pinecone, Weaviate, Qdrant, or pgvector, LangGraph / LangChain, LLM gateways or routing platforms, OpenTelemetry, LLM evaluation and tracing tools, AWS / GCP, GitOps / Argo CD, Model Context Protocol (MCP), experience building internal developer platforms.

• Berlin, Germany.