Platform Engineer (gn) @ AI Efficiency Venture, Berlin

This is an Atlantic portfolio venture. About the Venture AI models have moved faster than most companies’ ability to put them to work. The capabilities are there; knowing how to combine them effectively is still catching up. Frontier labs naturally put their own models first. Companies often follow suit, asking one general-purpose model to handle ten very different jobs. We believe there is far more to unlock in the models that already exist, even before the next generation arrives. That is the opportunity we are building around. We bring closed providers and open-weight models together behind a single API. Our engine breaks requests into their component parts, routes each to the right model within a cost budget, fine-tunes on customer data and checks the result against agreed quality standards. With no proprietary model to promote, we can choose the right combination for the task. Lower costs matter because of what they make possible: giving a small team the freedom to build something that would previously have required a much larger one. Our founder brings considerable drive, a hands-on approach and a strong sense of urgency to building the company. Backed by Atlantic at pre-seed, we are growing the engineering team to bring the product into production, with people already waiting to use it. About the Role As one of our first platform engineers, you will build the foundations that every product feature depends on: the gateway and router, usage and billing systems, customer key management and the European infrastructure underneath. Much of that platform is still taking shape. Working directly with the founder, CTO and founding AI engineer, you will help turn early architectural decisions into systems that remain reliable as the product grows. You will take capabilities from design through to production, thinking through their APIs, failure modes, deployment, observability and ongoing operation. In a team this small, building and running the system belong together. You will also work alongside our founding AI engineer on evals, embeddings and open-weight model serving, bringing platform engineering depth to the ML side of the product. Agentic engineering is central to how we work. You will use and help develop the harnesses and workflows that allow a small team to ship quickly, applying the coding fundamentals and judgement needed to stand behind what reaches production. There is a lot to build and an ambitious timeline ahead. We work with intensity, move quickly from decisions to execution and value people who take initiative and see things through. What You’ll Work On Your first area of ownership will depend on your strengths and our priorities. From there, you will contribute across the platform as it develops. Gateway and routing: build a consistent API across model providers, with reliable streaming, fallbacks and safe handling of retries and provider errors. Cost optimisation: develop caching, history compression and model selection, measuring savings while keeping estimated costs distinct from actual spend. Usage and billing: build accurate accounting for model calls, usage reports, plan entitlements and invoice reconciliation. Agent infrastructure and integrations: support tool calls, cost limits, human approvals and runs that can recover safely from interruptions, alongside connectors for services such as Gmail, Slack and calendars. Tenancy and secrets: maintain workspace isolation, secure authentication and encrypted customer key management. European infrastructure and model serving: build and operate our cloud infrastructure, including the GPU capacity used to serve open-weight models. Production reliability: make systems observable, migrations reversible and releases safe, with tests that protect data integrity and logs that keep customer content and secrets private. The platform will evolve quickly. You will need to move comfortably between improving what exists and learning enough about a ne