ML Performance Engineer: AI Acceleration & Systems

Tradermath; Greater London; Hybrid; GBP 100,000 - 150,000; Full time

Benefits offered by this job

Onsite gym, sauna, and fitness classes

Extensive medical benefits including a

Breakfast and lunch provided

Caregiver supports

Office Kings Cross location

25 days paid holiday

Generous Pension contributions

Job summary

XTX Markets in London seeks an experienced software engineer to boost ML performance and AI acceleration across training and inference platforms. You will work on a sophisticated optimizing compiler for accelerated computing, and push memory hierarchies, parallelism, and hardware-aware optimisations.

You should be proficient in C/C++/Rust and have experience with GPUs/TPUs, data-flow compilers, and OS-level development.

Qualifications

• Experienced software engineer in C/C++/Rust or similar language.
• Understanding of and experience targeting GPUs, TPUs, or similar hardware.
• Experience with data-flow oriented compilers for AI or scientific computing.
• Experience writing Linux kernel modules or other OS development.

Responsibilities

• Improve performance of XTX's ML training and inference platforms.
• Work on a sophisticated optimizing compiler for accelerated computing platforms.
• Analyze memory hierarchies, degrees of parallelism, and microarchitectural nuances to optimise code.
• Map AI models from JAX graphs to transistors and contribute to hardware architecture design.

Skills

C/C++/Rust

GPU/TPU experience

Data-flow compilers

Linux kernel modules

Tools

CUDA

JAX

Job description

XTX Markets in London seeks an experienced software engineer to boost ML performance and AI acceleration across training and inference platforms. You will work on a sophisticated optimizing compiler for accelerated computing, and push memory hierarchies, parallelism, and hardware-aware optimisations.

You should be proficient in C/C++/Rust and have experience with GPUs/TPUs, data-flow compilers, and OS-level development.