Intern - ML Inference Performance Engineer
Axelera AI12 hours ago
Eindhoven, NetherlandsIntern
Responsibilities
- Develop and improve benchmarking tools for throughput, latency, power, and accuracy across device-level, host-transaction, and end-to-end scenarios.
- Define reproducible benchmarking procedures, standardize results formats, and maintain a performance visualization dashboard.
- Evaluate AI accelerator products, SDKs, toolchains, model support, flexibility, and limitations across vendors.
- Characterize full inference pipelines, including host-device transaction overhead and end-to-end performance using equivalent configurations such as GStreamer.
- Set up and maintain lab hosts across multiple hardware platforms and onboard new evaluation hardware.
- Synthesize performance findings into structured reports that inform engineering and product roadmap decisions.
Requirements
- Currently enrolled in the final years of a Bachelor's program or in a Master's program in Computer Engineering, Electrical Engineering, Computer Science, or a related field.
- Python development experience and knowledge of C/C++.
- Experience with end-to-end computer vision pipelines and familiarity with benchmarking concepts such as performance and latency.
- Experience with inference tools, APIs, or SDKs such as TensorRT.
- Familiarity with deep learning concepts and technologies including quantization, ONNX, and PyTorch.
- Development experience using agentic AI and familiarity with LLM benchmarking concepts.
- Knowledge of Git and proficiency with Linux, Bash scripting, and Docker.
- Hands-on experience with embedded hosts.
- Strong written and verbal English communication skills, with the ability to document findings clearly and precisely.
- Good organizational skills.
- GStreamer knowledge and basic GUI design experience are nice-to-have qualifications.
Benefits
- Attractive compensation package including a pension plan, extensive employee insurances, and an option to receive company shares.
- Open, collaborative, inclusive culture supporting creativity, innovation, ownership, and freedom with responsibility.
- Work from Axelera AI's office in Eindhoven, Netherlands.
About Axelera AI
Axelera AI builds AI accelerators and an edge inference software stack (the Metis AI Platform) for computer vision and other workloads, selling chips, cards, and developer kits to OEMs and solution providers. Privately held and founded in 2021, it is headquartered in Eindhoven with R&D offices in Belgium, Switzerland, the UK, and Italy. The company focuses on on-device AI deployments where latency, power, and cost are critical.