AI / ML & DataRemote

Senior/Staff AI Application Engineer - Heterogeneous Computing

Axelera AI · Amsterdam, Bristol +2 (remote)

Axelera AI, headquartered in Eindhoven, makes the Metis AI platform and its own AI processing units. In the Model Enablement team, this senior or staff engineer decides how each model is split across the AIPU, host CPU and other accelerators, which drives most of the performance customers see in the Voyager SDK.

Apply on Axelera AI

Company: Axelera AI
Where: Amsterdam, Bristol +2 (remote)
Function: AI / ML & Data
Posted: 2 weeks ago
Team: AI Software
Work mode: Remote
Categories: AI & ML DeepTech

The role

What you’ll do

Build the model bring-up platform used by engineers and AI agents, with verification wired into CI
Own how PyTorch and ONNX models are captured, analysed and partitioned across device and host
Extend that split to large models, cascades and multi-model pipelines with the graph compiler team
Profile and tune end-to-end performance with the runtime and application teams

What you’ll need

Eight or more years in systems, compilers, runtimes or performance, including system-level design ownership
Hands-on heterogeneous computing across CPU, GPU, DSP, NPU or FPGA
Familiarity with ML graph capture, lowering, operator fusion and partitioning
Strong C++ and Python, and daily use of AI coding agents on real engineering work

Good to know: Remote or hybrid across Europe, with on-site options in several offices including Bristol. Priority goes to candidates based in, or willing to be based in, Belgium or Italy. Pension, insurances and share options.