Senior Embedded Performance Engineer

Description

The company was founded in 2022 on the idea that the speed of producing AI outputs, rather than model capability, would soon limit how much advanced AI can deliver. It designs chips and systems that generate output very quickly while running large models on very long contexts.

Today's most demanding AI workloads already deliver real value, but technical and economic limits on inference speed are holding progress back. Fields such as drug discovery, software engineering and materials research depend on long chains of reasoning, and faster inference would widen what they can achieve. The company recently raised $220M from investors including Founders Fund and Accel.

Responsibilities

• Write the on-device runtime software that schedules work across the chip and runs performance-critical ML kernels, shaping trade-offs for the silicon, deployment and compiler.
• Drive the first accelerator runs on silicon, measure performance, run early benchmarks and feed the results into the hardware and software roadmap.
• Write and tune performance-critical ML kernels in C, using assembly where it matters, for RISC-V and the company's own ISA.
• Build the low-level control code that feeds those kernels, covering scheduling, synchronisation and data movement.
• Create targeted validation workloads and microbenchmarks that keep simulation and hardware aligned and make performance measurable.
• Profile, benchmark and track regressions so that performance gains are real and repeatable.
• Work in a tight loop with simulation, hardware, ML, compiler, firmware and runtime engineers, turning profiling data into architecture feedback.

Requirements

• Proven experience with deeply embedded software.
• Strong performance intuition: you can reason about low-level architecture, memory behaviour and where cycles are spent.
• Expert in C, with a pragmatic approach to maintainable, high-quality low-level code.
• Able to write and debug optimised assembly, ideally for RISC-V.
• Collaborative and takes high ownership: you communicate clearly, move quickly and enjoy solving hard problems with others.
• A degree in Computer Science, Electronic Engineering, Maths, Physics or a related field, plus 3+ years of industry experience.
• Experience with GPUs or dedicated ML accelerators.
• Experience with Rust and/or Python.
• Experience with functional or performance simulators, and with writing validation or benchmarking workloads.
• Familiarity with modern ML inference workloads.

Conditions

Some roles involve technology subject to UK, US and other international export controls, and may require extra eligibility checks, which the company will explain during hiring.

The company is committed to equal employment opportunity and to building a diverse, inclusive team.

Benefits

• A competitive salary that reflects your experience and the specialist nature of the role.
• Meaningful equity so everyone shares in the value created.
• Private medical, dental and vision cover, a contributory pension, 25 days holiday plus bank holidays, and life and critical illness insurance.