About Fractile

Fractile was founded in 2022 on the bet that, eventually, the world’s most capable AI systems would be limited in their impact by the time taken to produce useful outputs. We bet everything on the logical conclusion: that the only way to truly unlock this latent value, to make speed viable at scale, was to radically re-invent the hardware that we run our frontier AI models on. Ever since, we have been building chips and systems that tackle this problem: how to efficiently generate output at thousands of tokens per second, while handling the complexity and capacity challenges of operating large models at very long contexts.

The workloads that push to the limits of the current frontier are already transformational; it is the technical and economic limits on inference speed that are constraining progress. The defining work of the 21st century will be marked by the engine of inference delivering immense and diffuse chains of intellectual inquiry, in drug discovery, in software engineering, in materials discovery, in any field where progress is driven by deep reasoning and intelligence to resolve complex problems.

About the Software organisation at Fractile

Developer Experience sits within the Software organisation at Fractile, which is responsible for developing a full software stack for our groundbreaking AI inference systems. That's everything from ML compilers, device drivers and systems firmware, application level runtime and ecosystem integrations, ML and compute libraries, great developer tooling and a full portfolio of simulators, through to datacenter scale workload deployment solutions. At Fractile, we know that a fantastic software stack is a critical and central part of any AI inference solution and it sits at the heart of everything we're doing.

About the team and role

The tools built by the Developer Experience team are how engineers — both inside Fractile and at our customers — see, control and understand the Fractile inference accelerator. That means the command-line utilities used to discover, configure and health-check devices; the diagnostics used during hardware bring-up and in the datacentre; and the profiling, tracing and debugging tools that show where the time goes across the host, the driver and the silicon.

As a Tools Developer you will design and build these tools in Rust, running on Linux hosts from a single developer workstation against large datacentre deployments. You will work closely with the hardware, firmware, kernel driver, runtime and Developer Experience teams to understand what they need, and turn that into fast, reliable and well-designed tools that fit naturally into existing Linux workflows. This is largely greenfield work, with real scope to shape how our tooling is architected and how it is used.

About you

You enjoy building tools that other engineers rely on every day, and you care about getting the details right: clear output, sensible defaults, good error messages and behaviour you can script against. You are at home on Linux, comfortable working anywhere from the shell down to sysfs, ioctls and the kernel/userspace boundary, and you know how to find out what a system is actually doing when it misbehaves. You are self-driven, able to identify tooling opportunities for yourself, and happy to work autonomously in areas of ambiguity.

Key Requirements

Strong Rust skills, with experience shipping production-quality tools or libraries

Knowledge of Linux from a developer's perspective: processes, memory, filesystems, sysfs/procfs, device files and ioctls, systemd and the kernel/userspace interface

Experience designing command-line tools and APIs that are easy to learn, script and automate

Experience working with stakeholders across multiple teams

Nice to Have

Experience writing tools for PCIe-attached devices, GPUs or other ML accelerators

Experience using CUDA, ROCm, Vulkan Compute, or similar

Familiarity with tracing and telemetry formats or systems such as Perfetto, Chrome trace format, OpenTelemetry or Prometheus