fig. 00 — α = 0.00

LinhaoMeng

I build data and interaction tools that help people understand, evaluate, and work alongside AI.

publications
9
first-authored
4
years building data & AI tools
8+
colour by

1,017 kinds of work people brought to Claude in Feb 2026. Slide α from what the work is to how people split it with AI: handing it over (warm) or working alongside it (cool). Dot size = conversations. Laid out with my Class‑Constrained t‑SNE.

Data: Anthropic Economic Index (2026-03 release, MIT) · topics clustered from task text · occupations: O*NET via the 2025-03 release · hover or tap a dot

fig. 01 · Now

Building AI tools people actually use.

Portrait of Linhao Meng
fig. 01a — the human in the loop

I am a software engineer at ByteDance, building developer tools for an AI model platform and creative tools where people and AI make content together.

Before that, I spent five years at Eindhoven University of Technology (PhD and postdoc) building algorithms and interactive tools that help people understand, compare, and debug AI models — and studying how they actually use them, from ML practitioners to clinicians.

fig. 02 · Work

Data, interaction, and the people in between.

I work where AI meets data and people: tools that let people inspect and evaluate the data and models behind AI, and interfaces that let them question, steer, and work with it. Interactive visualization and user studies are my main methods.

01

Data & evaluation tools

Tools for finding what is wrong in the data and models behind AI: hard instances, abnormal clients in federated learning, long-running training experiments, and side-by-side model comparison.

PhD & postdoc, TU/e · Oxford e-Research Centre · Zhejiang University
02

Interaction & human–AI collaboration

Interactive systems and user studies: how clinicians and ML practitioners use AI tools, what-if questions with counterfactuals, LLM-written explanations of data, and creative tools where people and AI make content together.

EU SmartCHANGE · ELLIIT, Linköping · ByteDance
03

Platforms & developer tooling

Developer tools for an AI model platform, and earlier, dashboards, a visualization component library, and open-source visualization for SynBioHub.

ByteDance · Cisco · Zhejiang University · Google Summer of Code

fig. 02a Selected projects

ModelWise overview figure
2022 · EuroVis 2022 / Computer Graphics Forum

ModelWise

Compare models, find why they fail

An interactive multi-model comparison tool that helps ML practitioners diagnose performance gaps, spot failure patterns, and choose models. Validated through user studies with practitioners.

First authorPaperCodeDemoVideo
Class-Constrained t‑SNE figure
2024 · IEEE TVCG / VIS 2023

Class-Constrained t‑SNE

Class-aware maps of ML datasets

A dimensionality reduction algorithm that combines data features with class probabilities, so a single 2D map shows both what the data looks like and what the model thinks — scaling to millions of points.

First authorPaperCode
VADAF figure
2021 · ACM TiiS

VADAF

Spotting abnormal clients in federated learning

A visual analytics tool for detecting and explaining abnormal clients in federated learning — without access to their private data.

First authorPaper
SynBioHub Sequence View figure
2021 · ACS Synthetic Biology (VisBOL2)

SynBioHub Sequence View

Google Summer of Code, NRNB

A JavaScript sequence visualization layer for DNA designs, following the SBOL Visual standard and integrated into the SynBioHub production platform.

Open-source contributorCodeWikiPaper
Visual Abstraction of OD Movement figure
2018 · IEEE TVCG / VAST 2018

Visual Abstraction of OD Movement

Millions of trips, one readable map

A geospatial visualization system that abstracts and renders millions of origin–destination movement records. Second author and core team member.

Second authorPaperVideo

fig. 02b Recent threads

IEEE VIS 2025 Short Paper

Difficulty-aware analysis of DNNs

Visual tools to find and characterize the hard instances in a dataset — the samples deep neural networks struggle with — pointing to targeted model improvements.

Under review

Glyphs for longitudinal ML experiments

A glyph-based visual analytics system for monitoring model pruning experiments over time, surfacing multi-metric trends across runs. Developed during a research visit to the Oxford e-Research Centre.

Postdoc, TU/e

Counterfactual & what-if analysis

Interactive tools for probing AI model behavior, built in the EU project SmartCHANGE together with clinicians and ML researchers.

VIS 2025 VISxGenAI Workshop

LangLasso

Integrating LLMs into visual analytics so that people get natural-language descriptions of the data clusters they select. A collaboration that began at the ELLIIT Focus Period in Sweden.

fig. 03 · Journey

From maps, to models, to makers.

Career timeline 2014–2026Study, research and industry roles over time; the full list is available below the chart.’15’16’17’18’19’20’21’22’23’24’25’26StudyResearchHighlightsIndustryB.S. GIScienceM.S.E.Ph.D.Research EngineerPhD ResearcherPostdocCiscoGSoCByteDanceOxfordELLIITPhD defensenow
Hover or tap a bar or markerTwelve years, four lanes.Geography → software → visual analytics → AI tooling.
  1. Mar 2026 – presentByteDance · Software Engineer, Data-AMLDeveloper tools for an AI model platform; human–AI creative tools
  2. Nov 2025PhD defensePhD thesis defended, Nov 2025
  3. Apr 2025ELLIITInvited Visiting Scholar, ELLIIT Focus Period, Linköping University, Apr–May 2025
  4. Apr 2025 – Dec 2025Postdoc · TU Eindhoven · EU SmartCHANGECounterfactual and what-if analysis tools; studying how clinicians use AI tools
  5. Jun 2024OxfordResearch visit, Oxford e-Research Centre, May–Jun 2024
  6. Sep 2020 – Apr 2025Ph.D. · TU EindhovenPh.D. in Computer Science, Visualization Cluster, supervised by Prof. Anna Vilanova
  7. Sep 2020 – Apr 2025PhD Researcher · TU EindhovenClass-Constrained t-SNE, ModelWise, glyph-based analytics, LangLasso, difficulty-aware analysis
  8. May 2020 – Aug 2020GSoC · NRNB / University of UtahOpen-source contributor: SynBioHub sequence visualization (VisBOL2)
  9. Aug 2019 – Apr 2020Cisco · Software Engineer Intern, DevNetProject-progress dashboard: React frontend, REST API, data model
  10. Sep 2018 – Jul 2020M.S.E. · Zhejiang UniversityM.S.E. in Software Engineering
  11. Feb 2018 – Jul 2020Research Engineer · ZJU Visual Analytics & Intelligence GroupWith Prof. Wei Chen: VADAF, OD movement visualization, banking dashboards
  12. Sep 2014 – Jul 2018B.S. GIScience · Zhejiang UniversityB.S. in Geographic Information Science
fig. 04 · Papers

Publications

10 shown · 9 published + 1 under review

  1. In review

    Glyph-based Visualization with Highlighting for Analyzing Longitudinal Machine Learning Experiments

    L. Meng, J. Chakraborty, S. van den Elzen, A. Vilanova, M. Chen

    Under review · Preprint on SSRN

    1st author
  2. 2025

    Towards Difficulty-Aware Analysis of Deep Neural Networks

    L. Meng, S. van den Elzen, A. Vilanova

    IEEE VIS Short Papers

    1st author
  3. 2025

    LangLasso: Interactive Cluster Descriptions through LLM Explanation

    R. Buchmüller, L. Meng, et al.

    VIS 2025 VISxGenAI Workshop

  4. 2024

    Class-Constrained t-SNE: Combining Data Features and Class Probabilities

    L. Meng, S. van den Elzen, N. Pezzotti, A. Vilanova

    IEEE TVCG 30(1) · VIS 2023

    1st author
  5. 2023

    Visual Reasoning for Uncertainty in Spatio-temporal Events of Historical Figures

    W. Zhang, S. Tan, S. Chen, L. Meng, T. Zhang, R. Zhu, W. Chen

    IEEE TVCG 29(6)

  6. 2022

    ModelWise: Interactive Model Comparison for Model Diagnosis, Improvement and Selection

    L. Meng, S. van den Elzen, A. Vilanova

    Computer Graphics Forum 41(3) · EuroVis 2022

    1st author
  7. 2021

    VADAF: Visualization for Abnormal Client Detection and Analysis in Federated Learning

    L. Meng, Y. Wei, R. Pan, S. Zhou, J. Zhang, W. Chen

    ACM TiiS 11(3–4)

    1st author
  8. 2021

    VisBOL2: Improving Web-Based Visualization for Synthetic Biology Designs

    B. Hatch, L. Meng, J. Mante, J. A. McLaughlin, J. Scott-Brown, C. J. Myers

    ACS Synthetic Biology 10(8)

  9. 2021

    Bubble Storytelling with Automated Animation: A Brexit Hashtag Activism Case Study

    N. Chotisarn, J. Lu, L. Ma, J. Xu, L. Meng, B. Lin, Y. Xu, X. Luo, W. Chen

    Journal of Visualization 24(1)

  10. 2018

    Visual Abstraction of Large Scale Geospatial Origin-Destination Movement Data

    Z. Zhou, L. Meng, C. Tang, Y. Zhao, Z. Guo, M. Hu, W. Chen

    IEEE TVCG · VAST 2018

Google Scholar profile

fig. 05 · Elsewhere

Visits, talks & tools of the trade.

Research visits

  • Apr–May 2025Invited Visiting Scholar, ELLIIT Focus Period on Visualization-Empowered Human-in-the-Loop AI
    Linköping University, Sweden
  • May–Jun 2024Research visit — visual design and analysis of model pruning experiments
    Oxford e-Research Centre, University of Oxford, UK

Talks & service

  • 2025Paper presentation — IEEE VIS · Vienna
  • 2023Paper presentation — IEEE VIS · Melbourne
  • 2022Paper presentation — EuroVis · Rome
  • 2022Poster — ICT Open
  • 2021Student volunteer — EuroVis

Awards

  • 2019Chinese National Scholarship
  • 20192nd Prize, ZJU AI Innovation Competition
  • 2016Zhejiang Provincial Scholarship

Data & evaluation

Model evaluation & diagnosisExperiment trackingExplainability (counterfactuals, what-if)Dimensionality reduction (t-SNE, UMAP, PCA)Data modeling

Interaction & research

User research & usability studiesInteractive visualizationDashboard designRapid prototyping

AI

LLM application developmentAgent workflowsTool / skill integrationModel fine-tuning

Build

TypeScript / JavaScriptPythonReactVueSvelteThree.js / WebGLD3.jsREST API design

Education

  • 2020–25Ph.D. in Computer Science
    Eindhoven University of Technology
  • 2018–20M.S.E. in Software Engineering
    Zhejiang University
  • 2014–18B.S. in Geographic Information Science
    Zhejiang University