LGLei Ge (雷鸽)Scientific AI · Computational FashionEmail

Curriculum vitae

Lei Ge (雷鸽)

PhD Candidate in Design Engineering at Imperial College London · LLM Machine Learning Engineer at Polaron · London, UK

Download PDF CV ↓
01

Current appointments

Imperial College London · London, UK

PhD Candidate in Design Engineering

  • Research on LLMs for materials science, interpretable AI, and agentic systems.
  • Supervisor: Dr Samuel J. Cooper.
Polaron · London, UK

LLM Machine Learning Engineer · Part-time

  • Leading the development of an LLM agent platform that bridges research and real-world deployment.
02

Education

Imperial College London · London, UK

PhD in Design Engineering

  • Research focus: LLMs for science, LLM agents, and AI interpretability.
  • Lee Family Scholarship, a full scholarship awarded to 1–2 top Chinese candidates annually.
Donghua University · Shanghai, China

MSc in Fashion Design and Engineering

  • GPA: 91/100; thesis grade: 92.6/100.
  • Thesis: A Computer Vision-Based Graphic Coding Technique for Garment Development.
  • Outstanding Master’s Thesis, China National Scholarship, Outstanding Student Award, and Shanghai Outstanding Graduate.
03

Research experience

Imperial College London

LLM surrogate belief under sparse observations

I study how language models behave as surrogate models in low-data optimization. The work shows that prompt design and query protocol are core components of surrogate specification, shaping uncertainty, belief updating, and downstream scientific decisions.

Imperial College London

LLM-guided optimization for battery charging protocols

I develop LLM agents for battery fast-charging optimization. The agents expand protocol search beyond fixed human parameterizations, incorporate natural-language safety constraints, and operate in expensive, slow-feedback scientific settings.

Imperial College London

Knowledge representation in large language models

I investigate how LLMs encode structured scientific knowledge. This work identifies a 3D spiral structure in hidden states that aligns with the conceptual organization of the periodic table.

Imperial College London

Automation and knowledge extraction for scientific data

I build automated systems for extracting, structuring, and operationalizing scientific knowledge from literature and imaging workflows, including MicroGPT for 3D microstructure analysis.

Donghua University

Vision-based 3D garment mapping and pattern development

I developed computer-vision methods for mapping flexible garment surfaces and translating 3D shape information into structured representations for apparel development.

Donghua University

AI-enabled personalized fashion design systems

I explored how individual body features and user-specific design needs can be translated into intelligent made-to-measure and apparel development workflows.

Donghua University

Zero-waste fashion pattern design

I investigated computational approaches to garment pattern-making that reduce waste while preserving fit, function, manufacturability, and design intent.

04

Publications

  1. P1

    Lei Ge, Samuel J. Cooper. Elicitation Matters: How Prompts and Query Protocols Shape LLM Surrogates under Sparse Observations . arXiv:2605.04764, 2026.

  2. P2

    Lei Ge, Ferran B. Planella, Stephen G. Baird, Samuel J. Cooper. From Prompt to Protocol: Fast Charging Batteries with Large Language Models . arXiv:2601.09626, 2026.

  3. P3

    Lei Ge, Samuel J. Cooper. Do Llamas Understand the Periodic Table? . Digital Discovery, 4(12), 3455–3465, 2025.

  4. P4

    Lei Ge, Ronan Docherty, Samuel J. Cooper. Materials Science in the Era of Large Language Models: A Perspective . Digital Discovery, 3(7), 1257–1272, 2024.

  5. P5

    Lei Ge, Xiaohui Li. A Vision-Based Approach to Mapping Flexible Objects for Garment Development . Textile Research Journal, 93(15–16), 3381–3395, 2023.

  6. P6

    Lei Ge, Xiaohui Li. A New Approach to 3D Pattern-Making for the Apparel Industry: Graphic Coding-Based Localization . Computers in Industry, 136, 103587, 2022.

  7. P7

    Lei Ge, Xiaohui Li. Review of Digital Pattern-Making Technology in Garment Production . Journal of Textile Research, 43(4), 203–209, 2022.

  8. P8

    Lei Ge, Xiaohui Li. A Pattern-Making Approach to Improving Zero-Waste Fashion Design . Fashion Practice, 13(3), 443–463, 2021.

05

Presentations

  1. T1
    Keynote

    From Trust to Action: Large Language Models for Scientific Discovery and Decision Support

    AI in Metallurgy (AIMET), Mechelen, Belgium

  2. T2
    Spotlight

    Do Llamas See the Periodic Table in 3D? Geometry and Layerwise Representations

    NeurIPS AI for Science Workshop, San Diego, USA

  3. T3
    Oral

    From Prompt to Protocol: Fast Charging Batteries with Large Language Models

    AI4AM, Madrid, Spain

  4. T4
    Oral

    Unveiling 3D Geometries in LLMs: The Representation and Recall of Periodic Elements

    AI4AM, San Sebastian, Spain

  5. T5
    Oral

    Large Language Models in Materials Science

    L2M3, EPFL, Switzerland

  6. T6
    Poster

    LLMs Know Chemistry: Representation of the Periodic Table in Large Language Models

    Oxford Modelling Symposium, Oxford, UK

  7. T7
    Poster

    Battery GPT? Materials Science in the Era of Large Language Models

    Oxford Modelling Symposium, Oxford, UK

  8. T8
    Poster

    Do Llamas Understand Chemistry?

    AI for Materials Summer School, Karlsruhe, Germany

06

Academic service

Google Cloud Startup Hub · London, UK

Co-organizer & Speaker · Deep Matters: Foundations

  • Co-organized an interdisciplinary conference on foundation models for materials science with approximately 100 attendees.
  • Worked across programme development, speaker coordination, sponsorship, logistics, and event delivery.
  • Secured sponsorship from Google and Polaron and coordinated invited speakers from NVIDIA, UCL, Orbital, CuspAI, and Imperial College London.
Website
MSCA Doctoral Network

Invited Lecturer · iCircular3 Marie Curie Doctoral Network

  • Delivered “How LLMs Work? From Transformer to Reasoning” for doctoral researchers.
Peer review

Journal reviewer

Reviewer for Communications Materials, npj Computational Materials, Digital Discovery, and RSC Advances.

07

Industry experience

Polaron · London, UK

LLM Machine Learning Engineer · Part-time

  • Built a production-oriented LLM orchestration service with FastAPI, LangGraph, and AWS Bedrock for natural-language control of internal workflows.
  • Designed multi-turn dialogue, streaming, tool execution loops, interruption and resume flows, and failure-recovery guardrails.
  • Implemented typed tool contracts, approval-aware workflow actions, sandboxed code execution, artifact handling, and retrieval over product documentation.
  • Added Redis, Postgres, and S3-backed persistence, checkpointing, audit logging, redacted tracing, readiness checks, and benchmark coverage.
Target · Shanghai, China

Management Trainee · Full-time

  • Completed rotations across Packaging, Color, Technical Design, Quality Assurance, and Sustainability.
  • Contributed to a sustainability initiative transforming recycled plastic bottles into home textiles.
Chenfeng Group · Changzhou, Wuxi & Suzhou, China

Programme Intern

  • Rotated across fashion design, intelligent sewing, and textile inspection to gain cross-functional exposure to apparel manufacturing.
H&M · Shanghai, China

Merchandising Assistant Intern

  • Supported merchandising workflows, created product forms and color charts, and facilitated supplier communication.
Target · Shanghai, China

Product Design and Development Intern

  • Designed zero-waste childrenswear for LTO Back to School, integrating sustainability across materials, manufacturing, and retail.
  • Supported own-brand development through market research, supplier visits, and certification feedback.