Mahmoud Mahfouz
AI Research Lead @ J.P. Morgan AI Research & PhD Student @ Imperial College London
I am an AI Research Lead at J.P. Morgan’s AI Research lab, an academic research lab led by Prof. Manuela Veloso, and a part-time Ph.D. student at Imperial College London, supervised by Prof. Danilo Mandic.
My research investigates the development of autonomous agents for complex, multi-agent financial market environments. I am interested in solving sequential decision-making problems in these dynamic, non-stationary, and partially observable environments. My work employs techniques such as deep reinforcement learning, behavioral cloning, agent-based modeling, and LLM-powered multi-agent systems. I primarily focus on applying these methods to (1) general decision-making problems in finance and (2)algorithmic trading problems within limit order book markets, covering areas like universe selection, asset allocation, and optimal trade scheduling and execution.
At J.P. Morgan, I began my career as a front-office software engineer developing and supporting global trader-facing software systems. I later transitioned into a data scientist role, designing machine learning models for time-series anomaly detection and anti-money laundering screening.
Before joining J.P. Morgan, I interned as a software engineer at Goldman Sachs, robotics engineer at Culham Centre for Fusion Energy (CCFE) and applications engineer at Keysight Technologies. My most interesting summer experience was definitly at the Culham Centre for Fusion Energy (CCFE), where I performed learnt a lot about the remote-handling robotic systems used in the Joint European Torus (JET) nuclear fusion tokamak.
I obtained my Master of Engineering (MEng) in Mechatronic Engineering from The University of Manchester, graduating with First Class (Hons) and ranked 3rd in the Electrical & Electronic Engineering department. During my studies, I built small and large robotic systems for industrial (electronics assembly) and research (landmine detection) purposes. My final year group project focused on (1) building a testbed for emulating landmine detection sweeping and (2) investigating the integration of various sensory data for speeding up the landmine detection algorithms developed by the group. The project was sponsored by Find A Better Way, a charity founded by Sir Bobby Charlton.