CUDA Developer

'

: - ( - )

Experience: 5+ yrs

Location: Bengaluru

Job Type: Full-time

We are looking for an experienced CUDA Software Engineer to develop and optimize high-performance GPU-accelerated applications for advanced image processing and computer vision workloads. This role is ideal for professionals with deep expertise in CUDA, C/C++, and GPU optimization who are passionate about building scalable, real-time processing pipelines.

As a CUDA Software Engineer, you will work on performance-critical software by optimizing compute-intensive algorithms, implementing highly parallel CUDA kernels, and improving execution efficiency across NVIDIA GPU platforms. You will collaborate closely with algorithm, software, and engineering teams to accelerate image processing pipelines while maintaining code quality, scalability, and reliability. This role offers an opportunity to solve complex computational challenges using modern GPU technologies and parallel programming techniques.

Key Responsibilities

• Profile, analyze, and optimize C/C++ and CUDA applications to achieve high-performance, real-time execution.
• Design, implement, and optimize CUDA kernels for image processing and computer vision algorithms.
• Improve GPU memory utilization, execution efficiency, and parallel processing performance across NVIDIA GPU platforms.
• Develop proof-of-concept solutions using Python for rapid algorithm validation and experimentation.
• Integrate GPU-accelerated components with existing software pipelines and image processing frameworks.
• Utilize GPU profiling and debugging tools to identify bottlenecks and optimize application performance.
• Collaborate with cross-functional engineering teams to translate algorithmic requirements into efficient GPU implementations.
• Write clean, maintainable, and well-documented C/C++, CUDA, and Python code following software engineering best practices.
• Validate functionality, optimize resource utilization, and ensure the reliability of GPU-accelerated applications.
• Contribute to continuous performance improvements and technical innovation across high-performance computing initiatives.

What Makes You a Great Fit

• 5+ years of hands-on experience in CUDA development and GPU performance optimization.
• Strong programming expertise in C/C++ with proficiency in Python for scripting and rapid prototyping.
• Deep understanding of NVIDIA GPU architecture, CUDA execution models, memory hierarchy, and parallel computing concepts.
• Experience using GPU profiling and performance analysis tools such as Nsight Systems, Nsight Compute, or similar utilities.
• Strong knowledge of OpenCV, including CUDA-accelerated modules and image processing workflows.
• Solid understanding of software optimization techniques for compute-intensive and real-time applications.
• Excellent analytical, debugging, and problem-solving skills with a strong focus on performance tuning.
• Ability to collaborate effectively with cross-functional engineering teams in a fast-paced development environment.
• Strong communication skills and commitment to delivering high-quality, scalable software solutions.
• A proactive mindset with a passion for GPU computing, parallel programming, and high-performance software engineering.