4 days ago
Responsibilities
- Develop innovative techniques for Generative AI training and inference, including transformer architectures, distributed parallelism, speculative decoding, and KV-caching strategies.
- Implement efficient Generative AI model architectures and demonstrate their benefits on AMD platforms.
- Integrate AMD-optimized models and libraries with PyTorch, JAX, vLLM, and SGLang, and publish training recipes.
- Collaborate with software and hardware teams to co-optimize end-to-end performance on current and future AMD solutions.
- Advance agentic workflows for optimizing and deploying Generative AI applications at scale.
- Publish and promote technical work at major conferences and collaborate with AMD, industry, and academic researchers.
Requirements
- Deep hands-on expertise in Generative AI model training and inference for areas such as LLMs, 3D world and action models, or image and video generation models.
- Expertise in efficient Generative AI algorithms, model architectures, optimized training, parallelism strategies, or low-precision training.
- Experience productizing Generative AI models and training foundation models at scale.
- Familiarity with PyTorch, JAX, vLLM, SGLang, and MuJoCo.
- Publications in relevant research areas are preferred, particularly at venues such as NeurIPS, CVPR, ECCV, ICCV, ICML, or ICLR.
- Several years of experience in AI, deep learning, and related software development.
- Excellent written, verbal, presentation, and internal and external coordination skills.
- PhD or master’s degree in computer science, electrical engineering, mathematics, or a related field.
Benefits
- Hybrid work arrangement in San Jose, California, with Seattle, Washington, or Austin, Texas as alternative locations.
- AMD benefits are offered as described in the company’s benefits information.
Tech Stack
Categories
AI ResearchML Engineering
About AMD
AMD designs and sells CPUs, GPUs, and adaptive/embedded computing products for PCs, data centers, gaming, and edge devices. Its portfolio includes Ryzen and EPYC processors, Radeon and Instinct graphics, and adaptive SoCs from its Xilinx acquisition, sold to OEMs, cloud providers, and device makers. Founded in 1969 and headquartered in Santa Clara, it is a public company on NASDAQ and supplies semi-custom chips for major game consoles.