AMD is looking for an influential software engineer role to enable AI acceleration at scale. You will be a member of the core team, working on developing ML tools and methodologies to optimize and realize full system performance for AI workloads on Ryzen AI SoC. Working on the latest AI models addressing vision, language, and generative models, working with the leading engineers in AMD s CPU, GPU, and Adaptable Compute team
KEY RESPONSIBILITIES :
- Design and develop efficient code-generation and optimization techniques for AMDs Machine Learning products using MLIR / LLVM
- Consistently research and implement methods to improve the performance of our solutions.
- Stay informed of software and hardware trends and innovations, especially pertaining to ML algorithms and architectures.
- Optimizing current system and research alternative, more efficient ways for same goals.
- Develop technical relationships with peers and partners.
- Work with AMD s architecture specialists to improve future products.
- Developing and optimizing code for VLIW processors.
PREFERRED EXPERIENCE :
Strong object-oriented programming background, C / C++ and / or Python with 3+ years of industry experiencePrior experience in Graph compilers and optimizationsDeep understanding of the performance implications on AI acceleration of different compute, memory, and communication configurations and hardware and softwareGood knowledge of AI frameworks like ONNX, Pytorch, TVM, TensorFlow,Familiarity with compiler technologies like MLIR / LLVM.Ability to write high quality code with a keen attention to detail.Familiar with implementation of key ML operations like GEMM, CONV.Experience with software development processes and tools such as debuggers, source code control systems (GitHub) and profilers.Effective communication and problem-solving skillsACADEMIC CREDENTIALS :
Bachelor s or Masters degree in Computer Science, Computer Engineering, ElectricalSkills Required
Machine Learning, C++, Artificial Intelligence, Soc, Python