As co-founder and CTO, you will help shape the company as well as its technology. Working alongside the founder as an eye-level partner, you will turn the ambition into an architecture, a product and an engineering team capable of delivering it.a

Your remit spans the efficiency engine, the platform beneath it and the people building both. You will lead our founding AI engineer and backend platform engineer, grow the team and establish the standards that guide its work. You have led engineering before, but stayed close to the code, and want to keep building as your responsibilities grow.

Two things define the technical challenge here. Efficiency is the product itself, so routing, fine-tuning, caching, quantization and open-weight serving sit at the heart of the roadmap. Agentic engineering is also central to how we work. You will set the standard for using individual agents and entire agent fleets effectively, building the harnesses and workflows that allow a small team to move quickly while retaining sound engineering judgement.

There is a lot to build and an ambitious timeline ahead. We work with relentless intensity, move quickly from decisions to execution and look for people who take initiative and see things through. You will have the scope to make consequential decisions early, and the responsibility to turn them into something that works.

You will move between architecture, code and company building, setting the technical direction while building the team that delivers it.

• Lead the engineering team: guide our founding AI engineer and backend platform engineer, review their work and establish the engineering and operational standards, including on-call and incident response.
• Build the team: lead technical hiring and decide how engineering grows as the company develops.
• Set the technical direction: make the architectural calls, shape the roadmap and decide what we build ourselves and where we use existing tools and services.
• Develop the efficiency engine: bring together routing, semantic and prefix caching, parameter-efficient fine-tuning, distillation, quantization and batching, with inference economics informing each decision.
• Own open-weight serving: choose and operate vLLM, SGLang or equivalent, including GPU capacity and European hosting, alongside closed providers through a single interface.
• Make quality measurable: build rigorous eval harnesses and the evidence needed to establish whether a cheaper route still meets the agreed standard.
• Build the platform underneath it: own the gateway API, provider layer, identity and tenancy, precise usage and cost accounting, secure storage of customer keys and European infrastructure as code, with GDPR and data residency requirements built in.
• Stay hands-on: write code and play an active part in designing and building the platform and cost-efficiency systems. Develop our agentic engineering setup, from harness design and benchmarking to the agent workflows that help the team ship.
• Represent the company technically: bring depth and clarity to customer pilots, enterprise security reviews and investor conversations.

You have grown into engineering leadership without losing the instinct to build. When faced with a difficult problem, you explore what would make it possible and look for a workable path through it, weighing failure modes alongside unit economics. AI has changed how you work, and you have taken the time to understand where it helps, where it fails and how to use it well.

• Engineering leadership: you have been a CTO, VP Engineering, Head of Engineering or engineering lead, ideally in a startup. You have hired and led engineers, established standards and taken responsibility for both the architecture and what the team shipped.
• Backend and platform engineering: you bring deep experience in distributed systems, API design, streaming, Postgres, Kubernetes, Terraform and cloud operations, with a broad understanding of what makes a platform reliable and scalable. You have taken systems into production, operated them under significant load and worked through incidents yourself.
• LLM cost efficiency: you have reduced costs in a production system and can explain the numbers, the decisions behind them and the trade-offs involved. Your experience covers techniques such as LoRA or QLoRA, routing, caching, quantization, distillation and batching, grounded in an understanding of serving economics.
• Open-weight models in production: you have served, tuned, evaluated and operated open models under load, including the GPU and capacity decisions that come with them.
• Agentic engineering: you build on strong coding fundamentals, with practical experience running agent fleets and designing and benchmarking their harnesses. You can reason independently about the code they produce and take responsibility for its quality.

You also bring a disciplined approach to LLM benchmarking: thoughtful eval design, statistical honesty and a willingness to reproduce results before relying on them.

Fluent English is required; German is a plus.

• A strong open-source record, particularly in AI infrastructure or inference, with real users.
• Experience with LiteLLM, NVIDIA NIM, TGI, llama.cpp or similar, and with self-hosted or air-gapped enterprise deployments.
• Experience with European cloud providers such as Scaleway, Hetzner or OVH.
• Experience working in startups tackling similar technical challenges.

Why Join

With the pre-seed round closed and the first steps already taken, you will join at a point where there is momentum to build on and considerable room to shape what comes next. You will build alongside the founder, choose and grow the engineering team, and receive co-founder equity in the business you help create.

The architecture, efficiency engine, security model and engineering culture are all still yours to shape. The decisions you make now will become the foundations for everything that follows.

We are entering a market where competitors have larger teams, more funding and an earlier start. Our conviction is that technical depth, focused decisions and a small team that works well together give us a meaningful opportunity to compete. For the right person, that is an interesting challenge to take on, with substantial ownership over how we approach it.

This is a full-time commitment, based in our Berlin office five days a week, starting as soon as possible.