Embedded Performance Engineer: ML Kernel Optimisation

Fractile; Greater London; Hybrid; GBP 50,000 - 70,000; Full time

Job summary

A tech company focusing on AI hardware is seeking an embedded software engineer in London or Bristol to optimize performance-critical ML kernels and develop low-level control paths. The role requires 3+ years of experience in embedded software, strong skills in C and assembly, and a collaborative mindset. You will profile and benchmark for performance improvements, working closely with hardware and software engineers. This position encourages fast-paced development with a focus on maintaining high-quality code.

Qualifications

• 3+ years of industry experience in embedded software.
• Deep understanding of low-level architecture and memory behaviour.
• Ability to build high-quality, maintainable low-level code.

Responsibilities

• Write and optimise performance-critical ML kernels in C.
• Build low-level control paths for scheduling and data movement.
• Profile and benchmark for performance improvements.

Skills

Embedded software experience

Performance instincts

Excellent C programming

Debugging optimised assembly

Collaborative communication

Education

Computer Science, Electronic Engineering, Maths, Physics, or related degree

Tools

RISC-V assembly

Simulation tools

Benchmarking tools

Job description

A tech company focusing on AI hardware is seeking an embedded software engineer in London or Bristol to optimize performance-critical ML kernels and develop low-level control paths. The role requires 3+ years of experience in embedded software, strong skills in C and assembly, and a collaborative mindset. You will profile and benchmark for performance improvements, working closely with hardware and software engineers. This position encourages fast-paced development with a focus on maintaining high-quality code.