Computer Vision/ML Software Engineer - GPU - 150921

Company: AMD
Location: Santa Clara, California, US
Site: https://jobs.amd.com/job/Santa-Clara…
Job Type: Onsite, Full-time

What you do at AMD changes everything

At AMD, we push the boundaries of what is possible. We believe in changing the world for the better by driving innovation in high-performance computing, graphics, and visualization technologies – building blocks for gaming, immersive platforms, and the data center.

Developing great technology takes more than talent: it takes amazing people who understand collaboration, respect, and who will go the “extra mile” to achieve unthinkable results. It takes people who have the passion and desire to disrupt the status quo, push boundaries, deliver innovation, and change the world. If you have this type of passion, we invite you to take a look at the opportunities available to come join our team.

Computer Vision/ML Software Engineer - GPU

The Role:

We are looking for our next team member to join our computer vision team in the Machine Learning Software Engineering group to enable and performance CPU and GPU code relating to computer vision. You will work independently and with other AMD engineers to tackle technical CPU and GPU performance issues and build test cases for verifying correctness and measure performance.

The Person:

Does this sound like you? We’d love to talk!

• Very Strong solution-oriented mindset
• Expertise in Computer vision, machine learning and performance testing on CPU & GPU
• Ability to independently prioritize opportunities to deliver results on time
• Excellent verbal and written communication skills
• Proficient in English

Key Responsibilities:

• Software design and development in C/C++ and Python
• Seek maximum performance on AMD GPU through a combination of performance optimization, workload characterization, compilers, math libraries and lower-level AMD-internal toolsets
• Feeding back performance bottlenecks to the relevant engineering groups to improve performance
• Collaborate on future architectures and performance testing
• Attend internal working groups in resolving engineering issues; contribute to the debug and testing of unreleased GPUs and their readiness for ML workloads
• Document and publish performance results and procedures you have generated and automate repeatable procedures
• Apply CV/ML for training/inference and deployment into production applications

Preferred Experience:

• Experience with computer vision and performance optimization
• Coding expertise in SIMD for CPU and GPU (C/C++, Python, OpenCL, HIP or Cuda)
• Performance profiling, monitoring tools, and software performance optimization
• Linux administration; understanding setup for ML middleware
• Familiar with computer vision algorithms and neural nets, such as Resnet, VGG, Yolo
• Experience with MLPerf or other ML benchmarks

• Experience working on very large NN models for computer vision
• Any experience understanding/inspecting/writing x86 assembly
• MPI and/or OMP coding experience
• Understanding of memory and cache hierarchy and methods to query performance/latency at each level
• Hands-on debugging skills with system-level debugger, as well as GPU debugging

Academic Credentials:

Bachelor’s degree in a technical field (Computer Science, Electrical Engineering, Physics, Mathematics), Preferred MS or PhD with machine learning coursework