Originally posted 23 October 2026 by the employer.
About the role
This role supports Master thesis students working on machine learning software for edge Neural Processing Units (NPUs). Projects focus on software areas such as speculative decoding, LLM inference via the Vulkan ML Emulation Layer, and memory- and performance-aware operator scheduling.
What you'll do
- Investigate speculative decoding methods for mixture-of-expert inference acceleration.
- Implement reliable and efficient sustained LLM inference utilizing the Vulkan ML Emulation Layer.
- Develop memory- and performance-aware operator scheduling strategies for Vulkan ML inference.
- Collaborate with Arm engineering teams in Lund to evaluate algorithms on target edge NPU architectures.
What you'll need
- Current enrollment in a Master's degree program affiliated with a Swedish university or college directly or via an exchange program.
- Understanding of machine learning, neural networks, and mathematics.
- Programming experience in C, C++, or Python.
- Familiarity with machine learning frameworks such as PyTorch.
- Professional proficiency in English.
- Right to work in Sweden for the full duration of the project without visa sponsorship.
Skills: NPU, Neural Networks, PyTorch, LLM Inference, Vulkan ML, C/C++
More semiconductor jobs: AI chip design jobs
This role has been open 1 day — well below the 87-day median for AI/ML Hardware Engineering roles.
AI/ML Hardware Engineering · AI ML Hardware
|
Open roles in category
506
|
Median days open
87 d
|
Median salary
$236k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | Arm Holdings | All employers we track in this specialty (506 roles · 76 employers) |
|---|---|---|
| Open roles in this specialty | 13 | 506 |
| Open roles in the wider Software, Firmware & Systems family | 130 | 6367 · 147 employers |
| Median days open | 1 d | 87 d (−86 d vs this employer) |
| Median salary (USD postings) | — | $236k |
Skills observed across this category: NPU, Neural Networks, PyTorch, LLM Inference, Vulkan ML, C/C++
Who's hiring in this category
- Qualcomm · 99 open roles · median 120 d
- NVIDIA · 89 open roles · median 62 d
- AMD · 32 open roles · median 112 d
- Micron Technology · 27 open roles · median 64 d
- Mobileye · 20 open roles · median 90 d
- Analog Devices · 17 open roles · median 47 d
How we counted: 506 open AI/ML Hardware Engineering (AI ML Hardware) roles from 76 employers tracked in the SemiconductorJobs index, counted 29 Oct 2026. Specialty figures count only roles carrying this exact specialty label, so an employer's related work in neighbouring specialties is not included there — it is counted in the wider Software, Firmware & Systems family row. Figures refresh nightly.