I am a Lead Data Scientist at Royal Bank of Canada (RBC) with over 7 years of experience in machine learning and data science, and an ML researcher at York University, where I conduct research at the Intelligent Visualization Lab.

In industry, I work across the full machine learning lifecycle, from business problem formulation and stakeholder collaboration to model development, deployment, evaluation, and monitoring. More recently, I have been focusing on multi-agent AI systems for financial applications.

My research focuses on data science agents, vision-language models for data visualization, chart understanding, model evaluation, and LLM post-training. I am particularly interested in building data science agents that can interact with real computer environments and autonomously complete end-to-end data science workflows.

Data Science Agents LLMs and VLMs Multimodal AI Reinforcement Learning Model Evaluation

News & Updates

Five papers were accepted to EMNLP 2026, including four first-authored papers, with three papers in the Main Conference, one in Findings, and one in the Industry Track.

One paper was accepted to ACL 2026 on evaluating the cross-lingual generalization of vision-language model judges.

Two papers were accepted to EACL 2026, including RL-Text2Vis, our multi-objective reinforcement learning framework for text-to-visualization.

The Perils of Chart Deception: How Misleading Visualizations Affect Vision-Language Models received a Best Short Paper Award at IEEE VIS 2025.

Three papers were accepted to EMNLP 2025, with two papers in the Main Conference and one in the Industry Track.

Our co-first-authored survey Review-based Recommender Systems: A Survey of Approaches, Challenges and Future Perspectives was accepted to ACM Computing Surveys, a leading journal with an impact factor of 30.4.

Two papers were accepted to ACL 2025, including work on chart question answering and vision-language model evaluation.

One paper was accepted to EMNLP 2024 Main Conference, and another paper was accepted to Computer Science Review.

One paper was accepted to ACL 2023, and another paper was accepted to EMNLP 2023.

I was accepted into PhD programs in Electrical and Computer Engineering at the University of Toronto, McGill University, and the University of British Columbia. I decided to decline the offers and pursue a career in data science and machine learning at RBC.

I graduated with a Master's degree from York University and published six papers from my research in peer-reviewed journals and conference proceedings.

Experience

Nov 2023 – Present

Lead Data Scientist

Royal Bank of Canada (RBC) · Toronto, Canada

Leading production machine learning, generative AI, and multi-agent systems for financial applications.

Jun 2021 – Oct 2023

Senior Data Scientist

Royal Bank of Canada (RBC) · Toronto, Canada

Developed and deployed machine learning and deep learning solutions across the full ML lifecycle.

2023 – Present

ML Researcher

York University · Toronto, Canada

Research on data science agents, vision-language models, multimodal reasoning, LLM post-training, and model evaluation.

Feb 2022 – May 2025

Instructor, Data Science and Analytics

School of Continuing Studies, University of Toronto · Toronto, Canada

Jan 2024 – Aug 2024

Contract Faculty, Object-Oriented Programming

York University · Toronto, Canada

Aug 2019 – Apr 2021

Data Scientist

Medisun · Chatham, Canada

May 2017 – Aug 2019

Research Analyst

HEATED Lab, York University · Toronto, Canada

Awards & Recognition

2026, 2022 RBC Quarterly Performance Award Performance Recognition
2025, 2022 RBC Quarterly Team Award Team Recognition
2021 McGill Engineering Doctoral Award (MEDA) $96K ($24K/year × 4) · Declined
2021 UBC Four Year Doctoral Fellowship $84K + Tuition Credit ($21K/year × 4) · Declined
2021 University of Toronto PhD Fellowship 4 Years Funding · Declined
2017–2019 York University Master's Scholarship $83K Total

Publications

Full publication list available on Google Scholar .

2026

Same Chart, Different Story: Bias in Vision-Language Chart Interpretation

EMNLP 2026

Beyond Static Charts: Can Language and Vision Language Models Generate Interactive Data Visualization Interfaces?

arXiv · 2026

VisEditBench: Can Vision-Language Models Edit Visualization Code from Multimodal Feedback?

EMNLP 2026

DSAgentBench: Can Agents Automate End-to-End Data-Science Workflows in Real Computer Environments?

EMNLP 2026

Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support

EMNLP 2026

BEADS: Bias Evaluation Across Domains

International Conference on Applications of Natural Language to Information Systems · 2026

Chart Deception in Vision-Language Models: From Vulnerability to Mitigation

arXiv · 2026

DATAREEL: Automated Data-Driven Video Story Generation with Animations

arXiv · 2026

Lost in Translation: Do LVLM Judges Generalize Across Languages?

ACL Findings · 2026

Aligning Text, Code, and Vision: A Multi-Objective Reinforcement Learning Framework for Text-to-Visualization

EACL · 2026

DashboardQA: Benchmarking Multimodal Agents for Question Answering on Interactive Dashboards

EACL Findings · 2026

2025

Deploying Tiny LVLM Judges for Real-World Evaluation of Chart Models: Lessons Learned and Best Practices

EMNLP Industry Track · 2025

LLM-Based Data Science Agents: A Survey of Capabilities, Challenges, and Future Directions

arXiv · 2025

The Perils of Chart Deception: How Misleading Visualizations Affect Vision-Language Models

IEEE VIS · 2025 · Best Short Paper Award

From Charts to Fair Narratives: Uncovering and Mitigating Geo-Economic Biases in Chart-to-Text

EMNLP · 2025

Text2Vis: A Challenging and Diverse Benchmark for Generating Multimodal Visualizations from Text

EMNLP · 2025

Evolution of ReID: From Early Methods to LLM Integration

arXiv · 2025

Judging the Judges: Can Large Vision-Language Models Fairly Evaluate Chart Comprehension and Reasoning?

ACL Industry Track · 2025

ChartQAPro: A More Diverse and Challenging Benchmark for Chart Question Answering

ACL Findings · 2025

2024

A Systematic Survey and Critical Review on Evaluating Large Language Models: Challenges, Limitations, and Recommendations

EMNLP · 2024

A Comprehensive Review of Recommender Systems: Transitioning from Theory to Practice

Computer Science Review · 2024

Review-based Recommender Systems: A Survey of Approaches, Challenges and Future Perspectives

ACM Computing Surveys · 2024/2025

FakeWatch: A Framework for Detecting Fake News to Ensure Credible Elections

Social Network Analysis and Mining · 2024

2023

Can Large Language Models Fix Data Annotation Errors? An Empirical Study Using Debatepedia for Query-Focused Text Summarization

EMNLP Findings · 2023

A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets

ACL Findings · 2023

CQSumDP: A ChatGPT-Annotated Resource for Query-Focused Abstractive Summarization Based on Debatepedia

arXiv · 2023

FakeWatch ElectionShield: A Benchmarking Framework to Detect Fake News for Credible US Elections

AAAI Workshop · 2023

Earlier

Anisotropic Thermal Conductivity of Inkjet-Printed 2D Crystal Films: Role of the Microstructure and Interfaces

Nanomaterials · 2022

High-Speed Contactless Sintering Characterization for Printed Electronics by Frequency-Domain Thermoreflectance

Flexible and Printed Electronics · 2020

Simultaneous Measurement of Anisotropic Thermal Conductivity and Thermal Boundary Conductance of 2-Dimensional Materials

Journal of Applied Physics · 2019

Effect of Sintering Conditions on the Thermal Properties of Printable Metal Nanoparticle Ink Studied by Thermoreflectance

SPIE Nanoengineering · 2019

Investigation of Anisotropic Heat Transport Through Frequency Domain Thermoreflectance Technique

York University · 2019

Anisotropic Thermal Conductivity of Printed Films Made of 2D-Materials Based Inks

APS March Meeting · 2019

Frequency-Domain Magneto-Optical Kerr Effect for Thermal Property Measurements of Anisotropic 2-Dimensional Materials

APS March Meeting · 2019

High-Frequency Measurements of Thermophysical Properties of Thin Films Using a Modified Broad-Band Frequency Domain Thermoreflectance Approach

Review of Scientific Instruments · 2018

Measuring the Thermal Properties of Anisotropic Materials Using Beam-Offset Frequency Domain Thermoreflectance

Journal of Applied Physics · 2018

Electron Transmission Through Non-Tunneling Regime of GaAs-AlGaAs Coupled Quantum Wells

GRIN Verlag · 2017

Non-Tunneling Regime of Coupled Quantum Well

American Academic Press · 2016