Graphcore Logo

Graphcore

PyTorch Engineer

Reposted Yesterday
Be an Early Applicant
Hybrid
Bristol, England
Senior level
Hybrid
Bristol, England
Senior level
Design, implement, optimise and maintain PyTorch framework support for Graphcore accelerators. Deliver features, tests, performance improvements, documentation and code reviews; collaborate across teams and upstream open-source projects; mentor colleagues and contribute to technical direction.
The summary above was generated by AI
About the job

Make PyTorch feel native, powerful and production-ready on Graphcore AI compute.

You will help build the software layer that connects Graphcore accelerators with state-of-the-art machine learning frameworks. Your work will shape how ML engineers and researchers use our hardware in real workloads.

You will design features, improve performance, maintain core code, and support users working at the frontier of AI. You will work deeply in PyTorch, with scope to contribute across other frameworks as the landscape evolves.

This is a role for engineers who enjoy hard technical problems and open-ended ownership. You will influence architecture, quality and direction while helping Graphcore unlock the next generation of AI breakthroughs.

The team and culture

The Frameworks team makes Graphcore hardware work seamlessly with the tools ML engineers already use, including PyTorch, Triton, JAX and TensorFlow. Work happens through agile practices, clear technical debate and close links with hardware, software and open-source teams.

You will join a Scrum team that owns features from design through delivery, performance, documentation and support. Engineers are expected to take responsibility, speak up early and make decisions grounded in evidence.

The team values pace without losing quality. You will be trusted to coordinate complex work, manage trade-offs and help others grow through reviews, mentoring and shared learning.

What we're looking for

· Strong software engineering skills, with the judgement to manage quality, complexity and delivery.
· Experience developing in Python.
· Experience building performant C++ applications in a commercial setting.
· Experience in at least one of ML kernels, compiler development, ML frameworks or computationally intensive engineering.
· Ability to coordinate complex technical tasks with cross-team dependencies.
· Clear communication, ownership and curiosity about how high-performance software works.

Benefits

· Unlimited annual leave
· Up to 5% matched pension
· Phantom equity – share in Graphcore’s success
· True flexibility in how and where you work
· Office spaces designed for collaboration
· Free food and an on-site barista
· Health cash plan
· Income protection
· Life assurance
· Along with other benefits you can choose from (private medical insurance, dental plan etc)

We welcome people from all backgrounds and experiences and are committed to building an inclusive environment where everyone can do their best work.

We’re an equal opportunity employer and recognise that everyone brings different strengths and perspectives. If you need any adjustments during the interview process, just let us know - we’re happy to support you.

Join the team at Graphcore

Graphcore is one of the world’s leading innovators in Artificial Intelligence compute. It is developing hardware, software and systems infrastructure that will unlock the next generation of AI breakthroughs and power the widespread adoption of AI solutions across every industry.

As part of the SoftBank Group, Graphcore is a member of an elite family of companies responsible for some of the world’s most transformative technologies. Together, they share a bold vision: to enable Artificial Super Intelligence and ensure its benefits are accessible to everyone.

Graphcore’s teams are drawn from diverse backgrounds and bring a broad range of skills and perspectives. A melting pot of AI research specialists, silicon designers, software engineers and systems architects, Graphcore brings together deep expertise to solve complex problems and deliver meaningful progress in AI compute.

If you want to help make advanced AI compute easier, faster and more useful for researchers and engineers, we’d love to hear from you. Apply now to be part of the journey.

 

Similar Jobs at Graphcore

4 Hours Ago
Hybrid
Entry level
Entry level
Artificial Intelligence • Semiconductor
Design, implement, optimize, and support PyTorch, Triton, and related framework features for Graphcore accelerators. Contribute to upstream PyTorch, maintain Python and C++ extensions, integrate compiler backend capabilities with torch.compile, review code, write documentation, and support users. Collaborate with software, AI/ML, research, and hardware teams while working in large open-source codebases.
Top Skills: C++GitGraphcore AcceleratorsPythonPyTorchTorch.CompileTriton
22 Minutes Ago
Hybrid
Senior level
Senior level
Artificial Intelligence • Semiconductor
Validate and benchmark ML models on Graphcore's software/hardware stack. Create automated benchmarking pipelines, run open-source models, design targeted tests for numerical precision, quantisation, attention, distributed execution, and debug performance and correctness issues across frameworks and environments.
Top Skills: Graphcore Software And Hardware StackJaxLinuxPythonPyTorchTensorFlowTriton
23 Minutes Ago
Hybrid
Senior level
Senior level
Artificial Intelligence • Semiconductor
Validate and benchmark ML models across Graphcore's software and hardware stack. Design and run automated benchmarking pipelines and targeted tests for numerical precision, quantisation, attention, distributed execution and model subgraphs. Debug model and system performance issues, expose regressions, and report findings to software, infrastructure and hardware teams to improve reliability and performance.
Top Skills: GraphcoreJaxLinuxPythonPyTorchTensorFlowTriton

What you need to know about the Edinburgh Tech Scene

From traditional pubs and centuries-old universities to sleek shopping malls and glass-paneled office buildings, Edinburgh's architecture reflects its unique blend of history and modernity. But the fusion of past and future isn't just visible in its buildings; it's also shaping the city's economy. Named the United Kingdom's leading technology ecosystem outside of London, Edinburgh plays host to major global companies like Apple and Adobe, as well as a growing number of innovative startups in fields like cybersecurity, finance and healthcare.

Sign up now Access later

Create Free Account

Please log in or sign up to report this job.

Create Free Account