Job Opportunities API

Embedded & Low-Level Software Engineer, Early Career Location: London or Bristol - Hybrid 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.  Fractile is seeking to increase the clock speed of global progress, one chip at a time. We’ve recently raised $220M from investors including Founders Fund and Accel and our most important work lies ahead. Join us! The Role In this role you’d work in a small, cross-functional area alongside engineers who have shipped kernel drivers, optimised code on real accelerators, built simulators, or brought up silicon before. They’re excellent at what they do and want you to be too. Depending on where your interests and strengths are, you might: Write and debug the Linux kernel driver (in Rust!) for our PCIe-attached accelerator: PCIe, DMA, interrupts and memory Write and optimise ML inference kernels in a high-level language and in assembly for RISC-V and our custom ISA, tuning them at the instruction level: performance software that runs at the hardware’s limits Build the firmware that keeps our hardware running safely and reliably: bare-metal and RTOS on board and device controllers (including the BMC), plus embedded Linux Build device security in software: the root of trust, secure boot, device identity, and firmware updates that stay safe end to end Build and extend our QEMU-based functional simulator, so software can be developed before the silicon arrives Debug across the hardware and software boundary: kernel crashes, race conditions, memory errors, device bring-up This is an early-career role: early in your low-level software career, not necessarily fresh out of university. Whether you’re in your first software job, a few years in, or moving into this kind of work from an adjacent field, if you’re excited to work close to the metal, we’d like to hear from you. Rather than pick a specialism up front, you apply once and we work out together which area fits you best. What we’re looking for The people who do well in these areas tend to think below the abstraction layers. If something’s slow or broken you want to know why, down to the register and the instruction, and you don’t stop until it works and you’ve measured it. You’ll likely have these foundations: Strong programming fundamentals in a systems language: Rust, C, or C++ Serious hands-on experience in Linux, bare-metal, or an RTOS Enough knowledge of at least one architecture to reason about what the hardware is doing If you’ve written GPU or ML kernels (CUDA, Triton, or similar), that maps directly onto our performance work and we’d like to see it. We get especially curious when you’ve already touched the kind of work these areas involve: a Linux kernel driver wrangling PCIe, DMA, and interrupts; firmware or embedded Linux on a board you helped bring up; a QE