The GPU Kernel Engineer will play a role at the forefront of optimizing performance for the machine learning software stacks, especially at training and inference, and support the team to develop new and cutting-edge systems. The ideal candidate will have a strong background in parallel computing, and hands-on experience in system level coding, debug methodologies, and large-scale machine learning experience.
This role focuses on CUDA kernel development and optimization. Distributed training experience is a plus.
Key Responsibilities
- Understand, analyze, profile, optimize, and provide guidance to the team on deep learning workloads on state-of-the-art hardware and software platforms to improve their efficiency with different levels of optimization
- Design and implement performance benchmarks and testing methodologies to evaluate application performance
- Build tools to automate workload analysis, workload optimization, and other critical workflows
- Triage system issues and identify bottleneck and inefficiencies by analyzing the sources of issues and the impact on hardware, network and propose solutions to enhance GPU utilization
- Support the team to develop appropriate kernels and systems for new model architectures and algorithms
- Participate in, or lead design reviews with peers and stakeholders to decide amongst available technologies.
- Review code developed by other developers and provide feedback to ensure best practices (e.g., style guidelines, checking code in, accuracy, testability, and efficiency).
- Contribute to existing documentation or educational content and adapt content based on product/program updates and user feedback.
- Represent MBZUAI at industry conferences and events, showcasing the institution’s cutting-edge HPC and deep learning capabilities and establishing MBZUAI as a global leader in AI research and innovation.
- Perform all other duties as reasonably directed by the line manager that are commensurate with these functional objectives.
- Validate CUDA kernel outputs and gradients against reference implementations, and benchmark representative shapes, dtypes, and model workloads.
Technical Qualifications
- Strong C++ skills and hands-on CUDA kernel development and optimization for deep-learning workloads.
- Understanding of GPU memory hierarchy, warp/block execution, and compute-memory trade-offs, with demonstrated profiling-driven optimization.
- Strong Python skills and experience integrating kernels with PyTorch or an equivalent framework, including numerical and gradient validation where needed. Nice-to-Haves:
- Experience with Triton, CUTLASS, or PTX/SASS analysis.
- Experience with multi-node distributed training or inference systems.
- Experience validating mixed-precision computations, such as BF16 or FP8.
Skills Required
- Strong C++ skills and hands-on CUDA kernel development and optimization for deep-learning workloads
- Understanding of GPU memory hierarchy, warp and block execution, compute-memory trade-offs, and profiling-driven optimization
- Strong Python skills and experience integrating kernels with PyTorch or an equivalent framework
- Experience with numerical and gradient validation
- Experience with Triton, CUTLASS, or PTX/SASS analysis
- Experience with multi-node distributed training or inference systems
- Experience validating mixed-precision computations such as BF16 or FP8
What We Do
First a passion, then an idea transformed into success – when it comes to pioneering automation and digitalisation technology, the ifm group is the ideal partner. Since its foundation in 1969, ifm has developed, produced and sold sensors, controllers, software and systems for industrial automation and for SAP-based solutions for supply chain management and shop floor integration worldwide. As one of the pioneers of Industry 4.0, ifm develops and implements consistent solutions to digitalise the entire value chain “from sensor to ERP”. Today, the second-generation family-run ifm group has more than 8,750 employees and is one of the worldwide market leaders. The group combines the internationality and innovative strength of a growing group of companies with the flexibility and close customer contact of a medium-sized company.
.png)
