2027 Summer Intern, MS/PhD, AI-driven ML Performance Engineering Intern
About this role
Engineers who enjoy exploring low-level machine learning optimization and autonomous driving hardware will thrive in this group. Waymo operates a specialized team that designs and implements the hardware, compiler, and software frameworks for custom silicon ML accelerators. Interns joining this group partner with industry leaders to design domain-specific language kernels and build automated generation harnesses.
The position exists to advance autonomous driving technology by optimizing performance and numerical accuracy on custom silicon. Candidates will spend their summer tackling complex kernel generation challenges using AI agents to remove the need for human involvement.
What you will do
- Collaborate with AI agents to build high performance low-level kernels in a domain-specific language for a custom machine learning accelerator.
- Verify the functional correctness and numerical accuracy of the developed kernels.
- Analyze kernel performance to pinpoint bottlenecks and determine necessary improvements.
- Build a custom harness to iteratively enhance kernel generation and eliminate manual intervention.
What you need
- Enrollment in a master's or doctoral degree program focusing on computer science, computer engineering, computational science, or a related quantitative field.
- Demonstrated experience applying AI agents and harnesses to boost coding performance.
- Working knowledge of essential transformer inference kernels, including performance analysis and optimization.
- Familiarity with at least one low-level machine learning accelerator domain-specific language such as CUDA, Triton, or Gluon.
Nice to have
- Solid understanding of modern inference accelerator architectures
- Fluency in Python and/or C++
- Working knowledge of different ML numerical formats and their relative accuracy vs performance tradeoffs
Benefits
- Generous benefits programs, subject to eligibility requirements
Skills and tech stack
Questions candidates ask
What is the work policy for this internship?
This will be a hybrid onsite internship position located in Mountain View, CA. Resumes are accepted on a rolling basis until the role is filled.
What is the compensation for the role?
The expected hourly rate is $70 for Masters students and $85 for PhD candidates. Interns are also eligible to participate in company benefits programs subject to eligibility requirements.
How should I apply if I am interested in multiple positions?
To be in consideration for multiple roles, you will need to apply to each one individually. Please apply to the top 3 roles you are interested in.