Silicon Infrastructure Engineer

Fractile is building the silicon, systems and software to break through the memory wall, the fundamental hardware constraint standing between today's AI and what comes next. 

The frontier of AI is no longer a research problem. The tasks AI can complete are doubling in complexity every six to seven months and the tokens required to complete them are scaling with it. Sequential reasoning, the kind that can't be parallelised away, means the internal clock speed of inference systems is the critical constraint.  What stands between where we are today and the future potential of AI isn't smarter algorithms; it's the hardware to run them fast enough to matter. 

Today's chips are hitting their wall. We're building the ones that don't. 
 
Fractile is seeking to increase the clock speed of global progress, one chip at a time. 

 Key Responsibilities:

• Create and support Python tooling and silicon verification and physical design workflows centred around silicon EDA tooling, which will require coding and build-system knowledge to assist with tasks faced by different teams.
• Bazel build system support for new silicon EDA tools required for throughout the chip lifecycle
• Create and improve Python developer tools across frontend and backend silicon teams
• Improve existing workflows to make them cacheable and reproducible
• Work with the engineering team to build and optimise their workloads

It would be great if you have:

• Experience of silicon EDA tooling
• Experience working with build systems tooling, such as Bazel.
• Experience working with workload management tools, such as Slurm.
• Experience working with container orchestration tools, such as Docker, and Kubernetes.
• Experience working with infrastructure as code, such as Ansible, or Terraform and maintaining compute infrastructure
• Experience of setup and monitor observation tooling for resource utilisation, machine failures, and more (e.g. Prometheus/Zabbix)

Preferred Qualifications:

• Proficient in modern software development language(s) especially Python
• Proficient dealing with modern build systems e.g. Bazel, Pants
• Past experience with diagnosing and resolving network/storage/CPU/RAM bottlenecks across complex workloads.
• Experience deploying and managing a grid compute system (Slurm/LSF/SGE).
• Proficiency with containerisation frameworks (Docker/Singularity)

  About us:

  • Founded 2022, we're 100+ people across London and Bristol, in the heart of the UK's frontier AI ecosystem, and growing fast. 

  • We offer competitive salaries, meaningful equity, and standard company benefits.

  • We believe the hardest problems get solved by the broadest range of minds. We actively encourage applications from underrepresented groups in hardware and software engineering. 

  • Hybrid working; 2-3 days in our London and Bristol offices. 

   Export controls:

  Our work involves technologies subject to UK and international export control regulations. Certain roles may require additional eligibility checks to ensure compliance with applicable law. We'll be transparent about this throughout the hiring process.