AI Platform Engineer at Picnic
Netherlands
Posted
on-site · full-time · Visa sponsorship
Responsibilities:
- Build and operate platform components that take AI from experiment to production: model access, routing, serving and fine-tuning pipelines.
- Develop an AI Gateway with provider integrations, model routing, fallbacks, rate limits, privacy controls and cost attribution.
- Implement tracing, evaluation and benchmarking across prompts, models, retrieval and agent state to measure outcomes.
- Productionise open-weight models: design fine-tuning, serving and inference infrastructure for both low-latency and high-throughput use cases.
- Provide developer-facing APIs, SDKs, templates and local tooling to simplify building, testing and deploying agents and LLM-based features.
- Collaborate with product, ML and platform teams to set technical direction, ensure reliability, security and cost-effectiveness.
Requirements:
- Bachelor’s degree or higher in Computer Science, AI or related field.
- Strong backend engineering fundamentals and daily Python experience (FastAPI or similar).
- Proven experience building self-service AI platforms or similar developer platforms and scaling adoption.
- Hands-on experience deploying LLMs/ML systems to production with attention to latency, reliability, cost and security trade-offs.
- Familiarity with cloud infrastructure, containers, Kubernetes, CI/CD, observability and infrastructure-as-code (Terraform, Helm).
- Able to take end-to-end ownership, make pragmatic build vs buy decisions and work effectively across platform and product teams.
- Bonus: experience with LLM gateways, tool protocols (e.g., MCP), guardrails, or LLM fine-tuning/serving ecosystems.