Jiawei Zheng is a applied machine learning researcher. He is currently working as a Postdoctoral Research Fellow in DigitLab. His recent work focuses on building LLM-based agents and environment in enterprise settings.

My research is driven by the goal of building intelligent systems to solve real problems at scale, which manifests itself in the following directions:

• LLM-based Agents: Exploring how language model-powered agents can reason, automate complex tasks, and support decision-making in enterprise settings. This includes developing consulting agents for eliciting digital transformation requirements.

• Human Behavioural Modelling: Understanding how people behave in their daily environments and how intelligent systems can capture and interpret these patterns meaningfully to enhance wellbeing, safety, and autonomy. This includes modelling activities, detecting anomalies, and forecasting behaviour, based on sensor collected sequential and video data.

• Process-aware Pattern Recognition: Investigating how contextual process (workflow) information from domain knowledge can help interpret patterns, bridging low-level features with high-level workflows. This includes consdiering uncertain events unfold over time, how they influence detecting deviations, and how to support context-aware automation.

• Blockchain and IoT: Exploring the intersection of IoT, Blockchain, and distributed computing, including secure data sharing and decentralised coordination mechanisms. This addresses the need for trustworthy, robust infrastructures for distributed AI systems operating across devices and organisations.

Prior to join DigitLab, he obtained PhD degree in AI from University of Edinburgh, supervised by Prof. Jacques Fleuriot, Prof. Jane Hillston and Dr. Petros Papapanagiotou. His PhD is funded by School of Informatics PhD Scholarship. Prior to that, he obtained MEng degree (National Scholarship) in Computer Science from Xidian University, supervised by Prof. Yulong Shen.