Day to day, that looks like: designing expert task creation flows, building task assignment and review interfaces, writing the data pipelines that move outputs from domain experts into structured training sets, and integrating LLM-powered tooling directly into the product. Some days it's product engineering. Some days it's closer to infrastructure. The common thread is that the software you build is what lets a small team produce high-quality evaluation data at scale.

You'll be the person who makes the platform real. Decide what gets built, how it gets built, and set the standard for how we build as we grow.

• Build and own the core data annotation and workflow platform end to end — from task assignment and expert interfaces to quality control and dataset delivery.

• Design systems that support expert onboarding, task routing, review workflows, and QA at scale.

• Build and improve AI-integrated pipelines across the platform, including LLM-assisted annotation, automated checking, and model-in-the-loop workflows.

• Work directly with founders and domain operators to turn manual, bespoke processes into reliable, scalable software.

• Make strong product and engineering decisions in ambiguous, fast-moving situations — scoping, prioritising, and shipping without waiting for perfect specs.

• Help define how we build as a team: tooling choices, engineering standards, and product direction.

You will not be a good fit if you thrive in well-defined scopes and prefer a clear spec before moving. We are looking for people who are comfortable identifying what needs building, making a call, and shipping — often before the full picture is clear.

You will not be a good fit if you like to go deep on one technical area for an extended period. We are looking for people who are energised by moving across product, infrastructure, data systems, and AI tooling — sometimes all in the same week.

You will not be a good fit if you prefer to hand work off at the boundaries of your role. We are looking for people who want to own problems end to end, from the first conversation through to something live.

You will not be a good fit if you aren't yet integrating AI into how you build day to day. We are looking for people who are already using LLMs and coding agents as part of how they work, and are excited to push that further.

You will not be a good fit if you think of speed and quality as things you trade off against each other. We are looking for people who don't cut corners — they cut scope.

We don't care about background, experience, or prestige. We want people who can demonstrate they will work hard, learn fast, and ship things that matter. Former founders, early engineers at startups, and people with infrastructure experience are a plus.

On-site in London. Compensation (salary + equity) will be competitive.