Machine Learning Hardware/Software Codesign Engineer

Job Title

Machine Learning Hardware/Software Codesign Engineer

Role Summary

Work on co-design of machine-learning software and accelerator/processor hardware for on-device inference. You will own significant portions of the software stack and prototypes, perform performance and power analysis, and collaborate closely with architecture, RTL, verification and physical-design teams.

Experience Level

Mid-level. The posting does not specify years of experience.

Responsibilities

Primary responsibilities include prototyping, analysis, and close collaboration with hardware teams. Expect hands-on development and performance evaluation work.

• Prototype and evaluate techniques to improve on-device inference without sacrificing accuracy.
• Perform comprehensive analyses of performance, power and accuracy from first principles and benchmarking.
• Build analytical models and implement prototypes to validate system optimizations.
• Optimize software for target platforms and assist in performance modeling and analysis of hardware/software interactions.
• Design and evolve software architecture and APIs for high-performance, extensible ML software.
• Port and optimize ML models and pipelines (e.g., PyTorch → intermediate representations/runtimes) for efficient inference.
• Provide requirements guidance and performance analysis to Architecture, RTL, Verification and Physical Design teams.
• Own and drive parts of the product-facing ML software and corresponding hardware co-design work.

Requirements

Key must-have skills and desirable extras.

• Must-have: Strong hands-on experience with Python and PyTorch for model development.
• Must-have: Experience with on-device inference and practical performance/power/accuracy tradeoffs and benchmarking.
• Must-have: Proficiency in C/C++ and software design for performance-sensitive systems.
• Must-have: Solid understanding of computer architecture (CPU/GPU), performance modeling and code optimization for platforms.
• Must-have: Strong communication skills and ability to work collaboratively on product-focused teams.
• Nice-to-have: Experience moving models from PyTorch into MLIR and optimizing inference with Triton, TensorRT, or similar runtimes.
• Nice-to-have: Background in DSP, signal processing or numerical acceleration techniques.

Education Requirements

Master's or PhD, or equivalent practical experience, in Computer Science, Computer Science & Engineering, Computer Engineering or a closely related technical field.

About the Company

Company: Akeana

Headquarters: Santa Clara, CA, USA

Akeana is a start-up specializing in processor design, with a team experienced in developing state-of-the-art products. Their expertise spans multiple architectures, including Risc-V, ARM, and x86, with backgrounds from prominent companies in the semiconductor industry. Funded by top-tier venture capitalists and industry leaders, Akeana focuses on innovative RTL design and micro-architecture development.