coding

Multimodal Agents: From Automation toward Open-Ended Self-Improvement

-
This thesis presents practical methodologies for building scalable multimodal agents that move from narrow automation toward open-ended self-improvement.

What Survives When Code Doesn’t?

-

B9 R2325

This talk explores how AI-driven code generation shifts the role of software from a durable artifact to a disposable implementation and argues for a new computational model for agentic software that formalizes the fundamental guarantees of intent, state, composition, and effect into explicit, enforceable contracts.

Making the silent genes speak

1 min read ·
Regulating an overlooked set of genes may help biologists manipulate cells to find new therapy options.

Learning about robots and coding through play

1 min read ·
"Hour of Code,” a worldwide initiative with the scope to teach children the basics of computer science, recently took place at The KAUST School Kindergarten. During Computer Science Education Week, December 5-9, 2021, researchers from the KAUST RISC Lab volunteered to help the children to “demystify” code and grasp their first concepts of robotics.

Mingchen Zhuge

Ph.D. Student, Computer Science
Mingchen Zhuge's research focuses on scalable multimodal agent systems, including code generation, agent swarms, agentic societies and economies, recursive self-improvement, open-ended evaluation, multimodal reasoning, and neural computers.