Mid-Level Python Developer (GenAI)

About the Role
We are looking for a GenAI Engineer to build and deliver practical AI and agentic features for client applications. You will implement multi-step agent workflows, connect LLMs with external APIs and databases, and take features from prototype to production. Working alongside international engineering squads, you'll help build production-grade AI applications that go beyond simple API wrappers.
What You Will Do:
Agent Workflows: Implement stateful agentic workflows, prompt templates, and tool-calling flows using frameworks like LangGraph.
System & Tool Integration: Connect LLMs and AI agents to internal and external APIs, databases, and third-party tools.
RAG Implementation: Build and optimize retrieval-augmented generation (RAG) pipelines using vector databases and structured outputs.
AI Safeguards: Implement standard safeguards for user-facing AI features, including rate limiting, content moderation, and basic guardrails.
Smart Engineering: Use AI-assisted development tools (e.g., Cursor, GitHub Copilot) to accelerate coding, write unit tests, and troubleshoot complex issues.
Evals & Quality: Apply prompt evaluation, tracing, and monitoring practices to ensure reliable output from AI components.
Collaborative Execution: Partner with backend engineers and QA to integrate GenAI components seamlessly into core product architectures.
What We Are Looking For:
Agentic Frameworks: Hands-on experience building agentic or LLM-driven workflows using frameworks like LangGraph, LangChain, or LlamaIndex.
LLM API Integration: Practical experience integrating LLM APIs and provider platforms (e.g., OpenAI, Anthropic, Google Vertex AI).
RAG & Vector Search: Working experience with vector databases (Pinecone, Weaviate, pgvector) and search/retrieval patterns.
Python & API: Strong async Python skills and experience building REST APIs with FastAPI to serve web or AI services.
AI Observability: Familiarity with AI guardrails, evaluation techniques, and observability tools (e.g., Langfuse, MLflow).
Communication: Professional English fluency; ability to collaborate effectively in an agile, remote-first environment.
Nice to Have:
Practical experience with AI coding tools (e.g., GitHub Copilot, Cursor), or openness to integrating them into your development habits.
Familiarity with tool-calling standards or protocols (e.g., MCP, A2A).
Hands-on experience with Docker, Kubernetes, and CI/CD pipelines (GitHub Actions, GitLab CI).
Familiarity with modern Python tooling (e.g., uv, Poetry, Ruff).
German language skills (B2+).

Employment type
Full-time

Work arrangement
No