MIT Grads Bring AI Expertise to IBM, Speeding Up Quantum Research
IBM has hired several recent MIT graduates to speed up its work in AI and quantum research. The team, led by PhDs Zhang-Wei Hong and Irene Ko, were part of the prestigious MIT-IBM Computing Research Lab.
Hong, who began working at IBM immediately after completing his doctorate in 2025, credits the lab with facilitating a seamless transition from academia to industry. He notes that the lab's academic collaboration policy and opportunity is better than other industrial labs.
Ko also benefited from her time at MIT-IBM Lab, where she started researching trustworthy AI alongside IBM researchers from day one of her PhD. Her work on vLLM inference engine plugins aims to provide cost savings by accessing internal model signals like hidden states and activations.
The team's research is focused on building robust and inquisitive AI systems that can self-evolve their weights during deployment. Hong believes this capability will be a major advancement for the field, enabling models to improve online at deployment time.