About the role#
As an AI-driven ML Performance Engineering Intern, you will work on the hardware, compiler, and software frameworks for our custom silicon ML accelerators. You will join a team that manages the full lifecycle of custom accelerator production, from initial architecture through to software integration.
What you'll do#
- Build high performance low-level kernels using a Domain-Specific language for custom ML accelerators.
- Verify the functional correctness and numerical accuracy of these kernels.
- Analyze kernel performance to identify bottlenecks and areas for improvement.
- Create a custom harness to automate kernel generation and reduce the need for manual intervention.
What you'll need#
- Enrollment in an MS or PhD program in Computer Science, Computer Engineering, Computational Science, or a related quantitative field.
- Experience using AI agents and harnesses to improve coding performance.
- Knowledge of transformer inference kernels, including optimization and performance analysis.
- Familiarity with low-level ML accelerator DSLs such as CUDA, Triton, or Gluon.
- Understanding of modern inference accelerator architectures.
- Proficiency in Python or C++.
- Knowledge of ML numerical formats and their accuracy versus performance tradeoffs.
Location & details#
- Location: Mountain View, California.
- Term: Summer 2027.
- Modality: Hybrid.
- Employment: Full-time.
- Compensation: $70 per hour for Masters students and $85 per hour for PhD students.
About Waymo
Waymo is a technology company that develops automated driving systems. It focuses on mobility and road safety through its Waymo Driver technology. The firm operates as a private organization and employs between 1,001 and 5,000 people. It was founded in 2009 and maintains its headquarters in Mountain View, California.
How to get in at Waymo
Applying early is a significant advantage for roles at Waymo, as recruiters often review candidates before the applicant pool grows too large. Intern Insider sends an instant alert the moment a role matching your target is published anywhere, so you can submit your materials among the first. This helps ensure your application is seen before the pile grows. You can also improve your response rates by reaching out to the people actually hiring for these positions. Intern Insider surfaces the recruiters behind the company roles, allowing you to contact them directly to ask about the team or a potential referral. It is a more effective approach than sending your resume into a queue.
