Fractile is a UK AI hardware startup building next-generation inference chips and systems for frontier AI.
<h1>Developer Experience Engineer</h1> <p>Fractile · London or Bristol · Full-time · Hybrid </p> <p class="font-claude-response-body break-words whitespace-normal"><strong>About Fractile</strong><br>Fractile is building silicon, systems, and software that redefine the frontier of AI — running the world's largest language models at radically higher speed and lower cost. Founded in 2022, our 100+ person team spans hardware and software, and we're growing fast.</p> <p class="font-claude-response-body break-words whitespace-normal">But revolutionary hardware only matters if developers can get their models running on it quickly and predictably. That's where you come in.</p> <p class="font-claude-response-body break-words whitespace-normal"><strong>The Role</strong><br>We're looking for Developer Experience Engineers to design and build the tooling and workflows that companies will use to optimise their large language models on Fractile's next-generation inference hardware.</p> <p class="font-claude-response-body break-words whitespace-normal">This is a role that sits at the intersection of deep systems knowledge and genuine empathy for developers. You'll shape how the outside world experiences Fractile — from a customer's first quickstart guide through to advanced performance profiling. You'll also be the voice of that external developer inside the company, turning real deployment pain into roadmap priorities.</p> <p class="font-claude-response-body break-words whitespace-normal">You'll work in a tight hardware-software co-design loop alongside ML compiler, runtime, systems software, and silicon engineers — influencing APIs and abstractions before they're set in stone.</p> <p class="font-claude-response-body break-words whitespace-normal"><strong>What You'll Do</strong></p> <ul class="[li_&]:mb-0 [li_&]:mt-1 [li_&]:gap-1 [&:not(:last-child)_ul]:pb-1 [&:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3"> <li class="font-claude-response-body whitespace-normal break-words pl-2">Design and build the tools customers use to run, optimise, and monitor large language models on Fractile hardware</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Build interfaces to diagnostic, profiling, and observability tools that help developers understand performance, debug issues, and optimise deployments</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Write documentation, quickstarts, reference examples, and tutorials that define a developer's first and lasting impression of our platform</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Act as the voice of the external developer inside Fractile — translating real deployment pain into roadmap priorities</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Collaborate with ML compiler, runtime, systems software, and silicon teams to shape APIs and abstractions early in the design process</li> </ul> <p class="font-claude-response-body break-words whitespace-normal"><strong>What We're Looking For</strong><br>We care about depth of knowledge, empathy for developers, and a genuine drive to make complex things feel simple. You'll be a strong fit if you have:</p> <ul class="[li_&]:mb-0 [li_&]:mt-1 [li_&]:gap-1 [&:not(:last-child)_ul]:pb-1 [&:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3"> <li class="font-claude-response-body whitespace-normal break-words pl-2">Strong software engineering fundamentals and experience building developer-facing tooling, SDKs, APIs, or platforms that other engineers depend on.</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Experience developing for inference hardware and/or GPUs (e.g. one year or more using CUDA or ROCm).</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">A good understanding of modern ML workloads and the practical challenges of deploying large language models to production.</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">An instinct for reducing the number of steps, surprises, and rough edges between a user and a working results.</li> </ul> <p class="font-claude-response-body break-words whitespace-normal"><strong>Bonus Points</strong><br>These aren't requirements, but they'd make you stand out:</p> <ul class="[li_&]:mb-0 [li_&]:mt-1 [li_&]:gap-1 [&:not(:last-child)_ul]:pb-1 [&:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3"> <li class="font-claude-response-body whitespace-normal break-words pl-2">Experience with inference serving stacks (vLLM, SGLang, TensorRT-LLM, Triton).</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">A track record of contributing to or maintaining open source developer tools and ML ecosystem projects</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Experience writing technical documentation and developer education content.</li> </ul> <p class="font-claude-response-body break-words whitespace-normal"><strong>How We Work</strong></p> <ul class="[li_&]:mb-0 [li_&]:mt-1 [li_&]:gap-1 [&:not(:last-child)_ul]:pb-1 [&:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3"> <li class="font-claude-response-body whitespace-normal break-words pl-2">Ownership and autonomy — you'll have real agency to drive your work forward</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Rapid iteration — we work directly with leadership to move from idea to product on fast-moving timelines</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Full-stack collaboration — hardware, software, silicon, and modelling teams work closely together</li> <li class="font-claude-response-body whitespace-normal break-words pl-2">Hybrid working: 3 days o