Eqvilent
С++ / СUDA developer
About this role
Join a dynamic team as a C++/CUDA developer to build high-performance GPU and CPU solutions with extreme latency and throughput demands. You'll implement complex computational algorithms, optimize existing systems for scalability, and work on cutting-edge low-latency projects in a fully remote environment.
What you'll do
- Implement complex computational algorithms on GPU and CPU with strict latency and throughput requirements
- Design and optimize custom CUDA kernels for performance
- Refactor existing solutions to improve scalability
- Profile and analyze application performance using specialized tools
- Debug high-performance GPU and CPU applications
- Collaborate with international team on project-based work
What they're looking for
- C++ (strong proficiency with data structures, algorithms, OOP)
- CUDA programming and custom kernel design
- GPU/CPU performance optimization
- Profiling tools (Nsight Systems, Nsight Compute, nvprof)
- Low-latency and real-time system development
- Linux system internals and networking
- Lock-free data structures
- Memory optimization techniques
Benefits
- Fully remote work from anywhere globally
- Access to global offices
- Flexible schedule
- 40 paid days off
- Competitive salary
- Cutting-edge hardware and technology
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Eqvilent
Eqvilent builds high-performance, low-latency trading infrastructure and systems that process massive financial market data volumes with extreme speed requirements. The company is hiring C++/CUDA developers, systems engineers, and software engineers to develop production trading platforms, optimize computational algorithms, and maintain globally distributed infrastructure.
View all jobs at EqvilentLikely interview questions
- Describe your experience designing and optimizing custom CUDA kernels. What specific challenges did you face regarding latency and throughput, and how did you address them?
- Walk us through your approach to profiling GPU applications. Which tools have you used (Nsight Systems, Nsight Compute, nvprof) and how did you use their results to optimize performance?