getSummary
Written in English from the original postingThis role involves building and optimizing low-level software and GPU kernels to run large machine learning workloads efficiently on specialized hardware. The engineer will deploy and operate multi-vendor compute clusters, production model-serving systems, and observability while engaging with customers. It is a full-time on-site position in San Francisco with a base salary of $200K-$300K and equity.
Job description
About the company
Our client is a technology company.
The role
Raydar is recruiting for this role on behalf of our client. Build and tune low-level software that makes large machine learning workloads run faster and more efficiently on a range of specialized hardware. Own production systems end to end, balancing reliability, observability and operating cost. Work closely with customers and engineers to shape the direction of the work.
What you'll do
- Bring newly released open-source models into production with performance tuned for speed and throughput.
- Refine the software that delivers model predictions, improving response times and how much work each machine handles.
- Write low-level GPU kernel code for accelerators from several hardware vendors.
- Design, deploy and operate compute clusters that mix hardware from different vendors.
- Operate live systems spanning varied hardware, watching uptime, monitoring and operating expense.
- Talk with customers about their needs, a meaningful share of the role.
Requirements
What we're looking for
- Evidence of exceptional results, such as top competition placement, a system owned end to end, or fast career progression.
- Backend, infrastructure or systems experience at a leading technology company or an early-stage startup.
- A deep understanding of computer architecture, memory hierarchy and operating system internals, beyond application-level coding.
- Able to start immediately and work on site in San Francisco.
- Enthusiasm for using AI agents and coding tools to improve their own workflow.
- Clear, warm communication skills for working with customers.
- A computer science degree from a top undergraduate program, with undergraduate study weighted above graduate study.
Bonus points
- Background in making machine learning models run faster and cheaper to serve.
- Track record keeping large, multi-machine hardware installations dependable and efficient.
- Exposure to GPU programming, ideally with a common accelerator toolkit.
- Open to new graduates with strong internships.
Benefits
Compensation and benefits
- Base salary: $200K - $300K
- Equity
- Health, dental and vision coverage
- Flexible time off
- Parental leave
Location and work model
- San Francisco, California, United States
- On-site, 5 days in-office
- Full-time