Job description for Lead GPU Kernel Developer at Telestar
ABOUT THE POSITION
You will develop novel algorithms with low-level kernels. You will collaborate closely with our algorithmic researchers to outperform existing SOTA kernels.
RESPONSIBILITIES
- Implement novel algorithms on GPUs, pushing hardware limitations to the edge.
- Proactively develop and implement novel speedup and automation methods
- Build and operate low-level profiling setups
- Stay up to date with hardware trends and new capabilities
- Communicate and collaborate with team members
OUR IDEAL CANDIDATE
Must have:
- Strong background in C/C++ and CUDA.
- Extensive experience in code profiling and performance optimization techniques.
- Outstanding problem-solving skills.
- Independent, quick learner.
Nice to have:
- Strong mathematical background.
- Deep understanding of AI algorithms.
- MSc/PhD in Math, CS, EE, or a related field.
BENEFITS
- Salary & Benefits are 100% Negotiable. We know top talent deserves a top-tier package. Whether your priority is base pay, a high-end equipment budget, or extra PTO—everything is on the table for the right candidate.
- Hybrid and WFH policy
- Opportunity to work in a dynamic and flexible environment.
- Regular team-building activities and company events.
- Social insurance and annual leave based on local labor laws.
- Year-end bonus based on individual and company performance.
