С++ / СUDA developer

The role

Implementation of complex computational algorithms on GPU and CPU with demanding latency and throughput requirements. Refactor existing solutions to improve their scalability.

What you will do

• Implementation of complex computational algorithms on GPU and CPU with demanding latency and throughput requirements
• Refactor existing solutions to improve their scalability

Required qualifications

• Commercial experience in developing and debugging high-performance GPU and CPU applications with strong focus on latency and throughput
• Hands-on experience with third-party libraries and designing custom CUDA kernels
• Proficient with profiling and performance analysis tools (Nsight Systems, Nsight Compute, nvprof)
• Solid understanding of data structures, algorithms, and object-oriented programming in C++
• Proven ability to work effectively in remote or hybrid teams with variable, project-based responsibilities
• Curiosity and proactive engagement with emerging trends in GPU/HPC/ML, continuously seeking to learn and apply new techniques

Nice to have

• In-depth knowledge and expertise in low-latency and real-time development, achieving sub-microsecond latency
• Experience with lock-free data structures and associated design patterns
• Proficient in C programming, including writing Linux drivers and working with kernel-level components
• Strong understanding of Linux system internals and networking
• Knowledge of memory optimization techniques (shared memory, registers, coalesced access, persistent kernels)
• Experience with inference optimization frameworks

Benefits

• Great challenges with fast feedback loops
• A welcoming group of highly qualified international professionals
• Cutting-edge hardware and technology
• Work remotely from anywhere in the world
• Access any of our global offices anytime
• Flexible schedule
• 40 paid days off
• Competitive salary

About the company

eqvilentjobs is a group of highly qualified international professionals.