About the Role
We are seeking an experienced engineering manager to lead the development of GPU-based machine learning accelerators. This role involves guiding technical direction, managing a team of engineers, and collaborating closely with research teams to optimize performance and scalability.
Responsibilities
- Lead and mentor a team of software engineers focused on GPU acceleration
- Define technical roadmaps for ML accelerator development
- Collaborate with machine learning researchers to implement efficient GPU kernels
- Drive performance optimization across large-scale training and inference workloads
- Oversee the design and implementation of low-level GPU code
- Ensure software reliability and scalability in production environments
- Coordinate cross-team efforts with systems and infrastructure groups
- Manage project timelines and deliverables for accelerator projects
- Evaluate emerging GPU architectures and programming models
- Promote best practices in software engineering and code quality
- Integrate accelerator components into broader ML training pipelines
- Troubleshoot complex performance bottlenecks in GPU workloads
- Contribute to hiring and professional development of team members
- Foster a culture of innovation and technical excellence
- Translate research concepts into production-grade software systems
- Work closely with hardware teams on co-design initiatives
- Maintain awareness of advancements in parallel computing and ML frameworks
- Ensure alignment between engineering output and research goals
- Drive adoption of new accelerator technologies across teams
- Balance short-term deliverables with long-term technical vision
Nice to Have
- Advanced degree in computer science or related technical field
- Prior experience managing engineering teams in AI or systems companies
- Contributions to open-source GPU software projects
- Hands-on work with AI accelerator chips beyond GPUs
- Experience optimizing kernels for throughput and latency
- Knowledge of formal verification methods for low-level code
- Background in compiler design for GPU targets
- Familiarity with safety-critical software development
- Experience with performance modeling and benchmarking
- Publications or patents in systems or ML acceleration
Compensation
Competitive salary and equity package
Work Arrangement
Hybrid or remote options available
Team
Part of the core systems team building advanced machine learning infrastructure
Research Collaboration
- Work directly with research scientists to bridge prototype and production systems
- Participate in early-stage research discussions to inform engineering priorities
Technical Depth
- Expected to contribute code and review low-level GPU implementations
- Lead design decisions for memory layout, kernel fusion, and data movement
