GPU Kernel Developer/AI Performance Engineer

Job Title : GPU Kernel Developer / AI Performance Engineer

Company: Wipro

Employment Type : Full-Time (FTE)

Experience : 4- 8 Years

Location : Hyderabad / Bangalore / Chennai / Pune / Remote

Job Summary :

learning models across different hardware platforms.

This role requires deep knowledge of GPU architecture, performance optimization, compiler/runtime execution layers, and low-level systems programming to improve AI model efficiency and scalability.

Key Responsibilities :

- Develop custom kernels using Triton for accelerating deep learning operations.

- Work on GEMM kernel development (matrix multiplication optimization).

- Port machine learning models to new hardware platforms and ensure compatibility.

- Optimize GPU compute performance by improving memory utilization, latency, and throughput.

- Perform system-level performance tuning, benchmarking, and stress testing.

- Work with compiler/runtime execution layers for model optimization.

- Develop and integrate custom SDKs or hardware abstraction layers.

- Collaborate with AI/ML teams to improve model inference and training performance.

- Troubleshoot low-level performance bottlenecks in GPU workloads.

Required Skills :

Programming :

- Strong expertise in Python

- Strong hands-on experience in C/C++

- Experience with PyTorch

- Hands-on expertise in Triton language/kernel development

GPU & System-Level Expertise :

- Strong understanding of GPU architecture

- Experience in compute optimization

- Experience with compiler optimizations/runtime execution layers

- Experience working with custom SDKs/HAL layers

Performance Optimization :

- Experience in ML model porting

- Strong experience in system-level performance tuning

- Stress testing and benchmarking expertise

Preferred Skills :

- Experience in AI acceleration frameworks

- Knowledge of deep learning model optimization techniques

- Experience with distributed systems or large-scale AI workloads