Recently, OpenAI has hired Ruoming Pang, a senior AI leader who most recently worked at Meta and previously led foundation model efforts at Apple, reflecting the escalating battle among tech giants to secure the engineers and researchers shaping the next generation of artificial intelligence systems.Pang’s career stretches across nearly two decades of major AI breakthroughs. His journey includes foundation models, infrastructure, and large-scale research that have evolved inside Silicon Valley’s most powerful companies.
OpenAI hires Ruoming Pang to advance its AI technology
In February 2026, Pang joined OpenAI as a Member of Technical Staff. The recruitment was reportedly months in the making.OpenAI’s interest in Pang appears aligned with its push to refine large language models, scale multimodal systems, and maintain a competitive edge in advanced AI research. Engineers who understand training pipelines, inference optimisation, distributed systems, and large-scale infrastructure are rare. Pang’s experience spans all of those domains.
Career at Google and contributions to large-scale AI infrastructure
Pang spent more than 15 years at Google as a Principal Software Engineer. His work covered multiple high-impact projects that quietly powered services used by billions.Within Google Brain, he helped lead speech recognition research and product integration, contributing to advancements that improved voice technology across Google’s ecosystem. He co-led development of the Babelfish and Lingvo deep-learning frameworks, which reportedly became some of the most heavily used training systems inside Google, particularly for TPU-based workloads.Earlier at Google, Pang co-founded Zanzibar, a globally consistent authorisation system that became critical infrastructure across thousands of internal projects. He also helped build structured search systems built on Bigtable infrastructure.
Ruoming Pang’s leading AI innovation and background
In 2021, Pang joined Apple as a Senior Distinguished Engineer, eventually leading the Apple Foundation Model team within Apple’s AI and machine learning division.His team developed the large-scale models that power Apple Intelligence, focusing on pre-training, post-training refinement, inference optimisation, and multimodal capabilities. He also led the development of AXLearn, a training framework designed to scale model training efficiently.At Apple, the work appears to have blended research ambition with product integration. Foundation models were not just theoretical experiments. They were engineered for deployment inside consumer products. That balance between research and shipping systems may have defined his leadership style.
Joined Meta to build future AI systems
In mid-2025, Pang moved to Meta as an AI Research Scientist, joining its Superintelligence Labs. The group was tasked with building the company’s next generation of advanced AI models. Reports suggested he joined with a compensation package valued at more than $200 million (1800 crore).His role reportedly involved overseeing AI infrastructure for Meta’s frontier research efforts. That likely meant scaling training systems, managing compute resources, and optimising model performance at a massive scale.
