Yu Wang Email & Phone Number

San Jose, California, United States

Artificial Intelligence and Data Center GPU Group product management

Palo Alto, California, United States

I played a pivotal role in driving the enablement of advanced compute workloads and GPU capabilities through VMware Cloud Foundation (VCF). I led the product definition, development and release of VMware’s first enterprise Generative AI solution, Private AI Foundation with NVIDIA, which empowered organizations to harness AI capabilities securely within their own data centers.My work involved close collaboration with cross-functional teams, including product management, engineering… Show more I played a pivotal role in driving the enablement of advanced compute workloads and GPU capabilities through VMware Cloud Foundation (VCF). I led the product definition, development and release of VMware’s first enterprise Generative AI solution, Private AI Foundation with NVIDIA, which empowered organizations to harness AI capabilities securely within their own data centers.My work involved close collaboration with cross-functional teams, including product management, engineering, marketing, sales, and customer success, to ensure that our solution met the evolving needs of the market. Show less

Product owner of VMware Private AI Foundation with NVIDIA.

As part of an internal rotation program ("Take3"), I was given the opportunity to be part of the vSphere product management team, and learn about the PM role in our organization.I was responsible for helping the CIBG PM team identify VMware's opportunities in capturing advanced workloads, and the MVP in new product features.

Palo Alto, California, United States

Vulkan Renderer team, Core Graphics, Cloud Infrastructure Business Group (CIBG)- Implementation in Vulkan backend to support graphics features to allow backwards compatibility for Workstation on existing graphics vendor platforms.- Working with our partner at AMD to enable VMware hosted products on AMD GPUs.- Finding root causes for host crashes, hangs, and rendering failures in AMD’s open source Vulkan driver (AMDVLK). Working with AMD partners to resolve these issues.

Palo Alto, California

GL Renderer team, Core Graphics, Cloud Platform Business Unit (CPBU)- Feature implementation in the backend to support HLSL Shader Model 4, Shader Model 5, and DX11.1 DDI, etc for Workstation, Fusion, and ESXi.- Investigating and fixing GL renderer regressions in Workstation and Fusion graphics stack.- Maintain test hosts running regression tests on Workstation and Fusion. Investigate, design minimum reproducible test cases, and report OpenGL driver regression to our partners… Show more GL Renderer team, Core Graphics, Cloud Platform Business Unit (CPBU)- Feature implementation in the backend to support HLSL Shader Model 4, Shader Model 5, and DX11.1 DDI, etc for Workstation, Fusion, and ESXi.- Investigating and fixing GL renderer regressions in Workstation and Fusion graphics stack.- Maintain test hosts running regression tests on Workstation and Fusion. Investigate, design minimum reproducible test cases, and report OpenGL driver regression to our partners (Apple, and NVIDIA)- Investigate, reproduce, debug and provide solutions to customer reported graphics issues. Show less

Palo Alto, California

* Optimize Windows OpenGL driver configuration on guest virtual machine. Design configuration file format and structure, implement the functionality in Windows driver and Mesa state tracker where extension and context configuration options are queried based on application process name before a context is created. (Language: C, XML parsing library: expat)* Design better Mesa GLSL compiler function inline heuristics. The criteria include function size (in IR instruction count), maximum call… Show more * Optimize Windows OpenGL driver configuration on guest virtual machine. Design configuration file format and structure, implement the functionality in Windows driver and Mesa state tracker where extension and context configuration options are queried based on application process name before a context is created. (Language: C, XML parsing library: expat)* Design better Mesa GLSL compiler function inline heuristics. The criteria include function size (in IR instruction count), maximum call depth, and function call frequency. Profile large GLSL shaders with complex call graphs. Integrate function inline heuristics in GLSL-to-TGSI opcodes conversion in the Mesa GLSL compiler implementation.* OpenGL extension: GL_ARB_clear_texture implementation in Mesa state tracker and svga driver Show less

Doctor Of Philosophy (Ph.D.)

M.S