Software Engineer Summer Intern, AI Compute Libraries & Performance 2027

<p><strong>Location:</strong> Bristol, UK<br><strong>Salary: </strong>£2500 per month<br><strong>Start date: </strong>June 2027<br><strong>Application deadline</strong>: Midnight, 29th November 2026 <br><em>We encourage you to apply early, as we may close applications before the advertised deadline.</em></p> <h3>The Internship Programme at Graphcore</h3> <p>Explore a career at Graphcore through a 12-week summer internship, gaining hands-on experience on real projects. You’ll bring fresh ideas, put your learning into practice and help shape what we’re creating. Along the way, you’ll tackle challenging problems with a team who welcome your questions and support you as you build your skills and confidence.</p> <h3>What you’ll work on</h3> <p>Use your programming skills and curiosity to help AI models run efficiently on next-generation Graphcore hardware. </p> <p>In AI Compute Libraries & Performance, you’ll contribute to C++ libraries and compute kernels for machine learning workloads. Your work will include: </p> <ul> <li><strong>Developing compute kernels:</strong> Implement and improve operations such as matrix multiplication (GEMM), convolutions, reductions, elementwise operations and fused kernels that combine multiple operations. </li> <li><strong>Understanding and improving performance:</strong> Use profiling and microbenchmarks to measure execution speed and investigate how threading, cache locality, memory layout and kernel launch efficiency affect performance. </li> <li><strong>Checking correctness and reliability:</strong> Build regression tests and numerical checks, investigate issues and improve the quality of the libraries our customers use. </li> </ul> <p>Through these projects, you’ll develop practical experience connecting numerical computing with how code runs on hardware, helping deliver software that is fast, accurate and reliable. </p> <h3>Your team and support</h3> <p>You’ll join a team building high-performance compute libraries that help customers get more from Graphcore hardware. Our work brings together machine learning, numerical computing and accelerator performance, with colleagues sharing their expertise to solve engineering problems.</p> <p><span data-contrast="none">You’ll contribute to clearly scoped projects and build your skills with guidance from engineers working in performance and low-level software. Together, we use profiling, experiments and technical reviews to compare approaches, understand trade-offs and improve our solutions.</span><span data-ccp-props="{"201341983":0,"335559738":240,"335559739":240,"335559740":240}"> </span></p> <p><span data-contrast="none">Your buddy will help you settle in, get familiar with our tools and ways of working, and find the right people to ask. Your manager will meet with you regularly to agree goals, offer feedback and discuss your development, including time for learning.</span><span data-ccp-props="{"201341983":0,"335559738":240,"335559739":240,"335559740":240}"> </span></p> <p><span data-contrast="none">We welcome different perspectives and make space for everyone to contribute. Questions, alternative approaches and requests for help are part of how we work. As you take on more responsibility, your colleagues will help you work through challenges, and your ideas will be listened to and discussed respectfully.</span><span data-ccp-props="{"201341983":0,"335559738":240,"335559739":240,"335559740":240}"> </span></p> <h3>How you’ll learn and grow</h3> <p>Over the 12 weeks, you’ll build performance engineering skills through hands-on work on compute kernels, benchmarks and tests.<span data-ccp-props="{"201341983":0,"335559739":0,"335559740":240}"> </span></p> <p><span data-contrast="auto">Profiling and experiments will help you understand where time is spent and how changes to code and data layout affect performance. Numerical validation and regression testing will show you how to check that an optimisation preserves correct results and continues to work as software changes.</span><span data-ccp-props="{"201341983":0,"335559