Dr Lilia Potseluyko

Name: Dr Lilia Potseluyko

Academic Division: Civil Engineering

Research Group: Construction Engineering – Digital Roads 

Email: lp625@cam.ac.uk

Personal Website:

linkedin.com/in/liliaamobi

liliapotseluyko.github.io

orcid.org/0000-0003-4135-8108

scholar.google.com/citations?user=AiCRbtUAAAAJ&hl=en

Dr Lilia Potseluyko is a Senior Industrial Researcher in Digital Twins at the Division of Civil Engineering, University of Cambridge. Working with Costain, National Highways and the Department for Transport, she leads technical discovery, user-centred product design, and development of data-driven decision-support tools for road maintenance and safety.

Her current work includes RoadGP, a digital-twin prototype that brings together road imagery, defect records, repair history, and AI-assisted recommendations. Lilia works with inspectors, engineers, contractors, and public-sector stakeholders to understand decisions and workflows, then translates them into user journeys, product requirements, interface concepts, and system architecture. Her delivery work spans Figma prototypes, Information architectures, Python services, REST APIs, geospatial data, and computer-vision pipelines.

Before joining Cambridge, Lilia led client-facing Digital Twin and automated schematic-mapping products at Nicander, taking concepts from technical discovery through interactive 3D interfaces and operational deployment for Swedish tunnel infrastructure. The products led to follow-on commercial projects. Earlier, at Norscot Joinery, she helped embed BIM, XR and game-engine workflows, earning an Innovate UK KTP Outstanding grade and a 2020 Scottish Innovator of the Future commendation.

Her PhD in Construction and Data Science specialises in BIM and immersive decision-support applications for the offsite construction sector. She holds an MA in 3D Design for Virtual Reality Environments from Glasgow Caledonian University and a Bachelor’s degree from Khabarovsk State Academy of Economics and Law.

Research Interests

Lilia’s current research interests include agentic AI and the design of robust multi-agent systems for infrastructure decision support. She is exploring retrieval-augmented generation (RAG), knowledge management, tool-using agents, agent orchestration, state management, evaluation and guardrails. Her focus is on connecting language models with road data, imagery, asset records and analytical models through Google Cloud services, while retaining human oversight to create transparent, reliable and scalable tools for road maintenance and safety.