AI Platform Engineer (AI Start-up)
First seen by Alion on Oct 14, 2026.
Rockstar Recruiting is hiring on behalf of a fast-growing, well-funded AI start-up, backed by one of Europe's top venture capital firms, that is building an AI operations platform for large enterprises, with customers among Europe's biggest retailers. This is a senior AI Platform Engineer role for someone who has shipped AI systems in production and wants to own the infrastructure behind them: agent runtime, evaluation pipelines, context management, tool orchestration and observability, using Python and TypeScript on GCP.
The team works from small hubs across Europe rather than fully remote, so the role is open to candidates based in, or ready to relocate to Berlin, Lisbon, or London, with the right to work in the EU or UK. Please get in touch with Tijana of Rockstar Recruiting to discuss further details: [email protected]
Requirements
- Experience building and running AI systems in production (not research): LLM-based applications or agents used by real customers
- Strong system design for AI-specific challenges, such as non-deterministic outputs, latency and cost optimisation
- Experience building evaluation frameworks to measure and improve the quality of AI systems
- An ownership mentality: you ship solutions to ambiguous AI problems and debug them in a structured, hypothesis-driven way
- Based in, or ready to relocate to, Berlin, Lisbon, or London , with the right to work in the EU or UK (no visa sponsorship)
Responsibilities
You'll join a fast-growing, well-funded AI start-up backed by one of Europe's top venture capital firms. They're building an AI operations platform for large enterprises, currently focused on retail and consumer goods. Their AI agents carry out business-critical work across SAP, spreadsheets, supplier portals, email and APIs, so the agents have to be reliable, fast and safe in production.
What you'll do
You'll own the AI infrastructure layer that makes the agents work in production:
- Build and improve the agent runtime, including tool orchestration and sandboxed execution
- Design context management systems that give agents the right information at the right time
- Build evaluation pipelines to measure agent quality and catch regressions before customers do
- Create observability tooling to understand agent behaviour, failures, latency and cost
- Own prompt engineering for core system behaviours, and make calls where benchmarks are still incomplete
This role is about AI infrastructure. It's not an ML research role, and not a role focused on building user-facing features or UI.
The team
The engineering team has around 20 people and is growing. About half are based at the home base in Prague, and the rest work from small hubs across Europe, all within two hours of Prague time.