AI Engineer – Applied LLMs, Workflows & Evals
Delvo
Seniority
Midweight
Model
In-Office
Sector
Salary
Undisclosed
Contract
Full-Time
About the role
Build reliable LLM workflows and agent systems for Delvo's procurement intelligence platform. You'll design production-ready AI systems with strong evaluation frameworks, moving beyond demos to real enterprise environments that deliver measurable results.
What you'll do
- Develop LLM workflows for retrieval, tool use, structured outputs, and multi-step reasoning
- Build agent orchestration systems with tools, control flows, retries, and safety checks
- Create evaluation and monitoring infrastructure including golden datasets and regression detection
- Implement enterprise integrations with ERPs and data sources with strong observability
- Optimize cost and latency through caching, streaming, batching, and intelligent model routing
- Collaborate with design and forward-deployed engineers to ship AI for real users
What you'll need
- Strong proficiency in TypeScript and Python for backend and product work
- Hands-on experience with LLMs including RAG, function calling, structured parsing, and guardrails
- Experience with modern AI stack: Vercel AI SDK, OpenAI/Azure, embeddings, vector stores
- Evaluation and quality mindset with ability to define tasks, gold data, and success criteria
- Systematic approach to preventing regressions and maintaining AI system quality
Nice to have
- Experience with data pipelines and ERP integrations
- Knowledge of procurement domain
- Experience with Langfuse or similar monitoring tools
What they offer
- Competitive salary plus meaningful equity with real ownership opportunity
- AI-first tooling with unlimited tokens and best-in-class development tools
- Deep technical growth in agentic systems and production AI
- Direct founder access in a small team environment
- Office located in CIC Berlin, Kreuzberg startup ecosystem
- Path to AI/ML leadership as the company scales