- Develop parallel mining algorithms and models that scale up to multiple GPUs and instances
- Identify performance hotspots and bounds with CUDA profiling tools
- Perform in-depth code analysis and parallel model optimization to exploit the potential of latest GPU architectures and maximize CPU/GPU utility
- Hands on experience with CUDA C++ parallel programming, including CPU/GPU integration for high-throughput systems
- Sound knowledge of different generations of CUDA architectures, compute capabilities, performance optimization techniques
- Solid knowledge of software design and programming techniques
- Basic understanding of algorithms and mathematics
- Application - Resume review and email exchange.
- 1st round interview - Technical review.
- Work sample (optional)
- 2nd round interview - Technical background interview with corporate partners.
Upon successful interview process, we will make an offer to the candidate and discuss on boarding package.