AI Hardware Acceleration Engineer for ML Performance
XTX Markets, Whitehall, City of Westminster
AI Hardware Acceleration Engineer for ML Performance
Salary not available. View on company website.
XTX Markets, Whitehall, City of Westminster
- Full time
- Permanent
- Onsite working
Posted 1 week ago, 4 Jun | Get your application in now before you're too late!
Closing date: Closing date not specified
Job ref: 6f55cd0da3544220a0c73b04919bdf64
Location ref: Whitehall, City of Westminster
Full Job Description
AI Hardware Acceleration Engineer for ML Performance
Date posted: 2026-06-03T00:00:00
Location: Greater London, London, UK, WC2N 5DN
Apply by: 2026-09-01T09:13:11.863
Company: XTX Markets
Role overview: XTX Markets, a leading trading firm in London, is seeking an experienced engineer for its ML Performance and AI Acceleration function. The successful candidate will join a small, highly technical team to enhance the performance of training and inference platforms. This role involves complex algorithmic challenges and optimizing compilers for advanced computing hardware. Candidates should possess strong skills in C/C++/Rust and have experience with GPUs, TPUs, or data-flow oriented compilers. The position offers exciting challenges and an innovative work environment.
- Strong skills in C, C++, Rust
- Experience with GPUs, TPUs, or data-flow oriented compilers
- Experience with optimizing compilers for advanced computing hardware
Job ID: #J-18808-Ljbffr
#s1-Gen
XTX Markets, a leading trading firm in London, is seeking an experienced engineer for its ML Performance and AI Acceleration function. The successful candidate will join a small, highly technical team to enhance the performance of training and inference platforms.This role involves complex algorithmic challenges and optimizing compilers for advanced computing hardware. Candidates should possess strong skills in C/C++/Rust and have experience with GPUs, TPUs, or data-flow oriented compilers.The position offers exciting challenges and an innovative work environment.