About News Publications Projects Demos Experience Open Source Contact
Dennis Johan Loevlie

Dennis Johan Loevlie

|

Download CV

I am an ELLIS PhD student at CWI and the University of Amsterdam, working in the Table Representation Learning Lab (TRLab) and the Amsterdam Machine Learning Lab (AMLab). I am advised by Madelon Hulsebos (TRLab, CWI) and co-advised by Jan-Willem van de Meent (AMLab, UvA). My research focuses on multimodal tabular foundation models.

Previously, I completed my M.S. in Computer Science at Tufts University, where I was advised by Michael Hughes on deep multiple instance learning for computational pathology. I also hold an M.S. in Chemical Engineering from Carnegie Mellon University and a B.S. in Chemical Engineering from West Virginia University.

Interests

  • Generative AI
  • Multimodal Foundation Models
  • Tabular Representation Learning
  • Data-Efficient Learning

News

2026 Won 1st place out of 10 teams in the BrainStorm Neural Decoder Challenge — real-time auditory decoding from 1024-channel ECoG with 95% accuracy and sub-ms inference.
2026 Starting as an ELLIS PhD student at CWI & University of Amsterdam in the Table Representation Learning Lab!
2025 Our paper "Synthetic Data Reveals Generalization Gaps in Correlated Multiple Instance Learning" accepted at ML4H 2025 Findings Track! Poster presentation in San Diego, December 2025.
2024 Awarded Community Grant from Hugging Face to demonstrate Depth Anything results on videos.

Publications

Selected Publications

Synthetic Data Reveals Generalization Gaps in Correlated Multiple Instance Learning
conference

Synthetic Data Reveals Generalization Gaps in Correlated Multiple Instance Learning

Ethan Harvey, Dennis Johan Loevlie, Michael C. Hughes

ML4H 2025 Symposium, Findings Track, 2025

Paper Code
Demystifying the Chemical Ordering of Multimetallic Nanoparticles Cover art by Sungil Hong & Dennis Loevlie
journal

Demystifying the Chemical Ordering of Multimetallic Nanoparticles

Dennis Johan Loevlie, Brenno Ferreira, Giannis Mpourmpakis

Accounts of Chemical Research, 2023

Paper Code

Other Publications

2023 journal

Single Atom Alloys Segregation in the Presence of Ligands

Maya Salem, Dennis J. Loevlie, Giannis Mpourmpakis

The Journal of Physical Chemistry C, 2023

Paper
2022 journal

Size-Dependent Shape Distributions of Platinum Nanoparticles

Ruikang Ding, Ingrid M. Padilla Espinosa, Dennis Loevlie, Soodabeh Azadehranjbar, Andrew J. Baker, Giannis Mpourmpakis, Ashlie Martini, Tevis D. B. Jacobs

Nanoscale Advances, 2022

Paper
2022 journal

Resolving Electrocatalytic Imprecision in Atomically Precise Metal Nanoclusters

Anantha V. Nagarajan, Dennis Johan Loevlie, Michael J. Cowan, Giannis Mpourmpakis

Current Opinion in Chemical Engineering, 2022

Paper
2020 poster

Software Development for HER High-Throughput Experiments

Dennis Loevlie

CMU ChemE Masters Student Association Research Forum, 2020

Poster
2018 poster

Mathematical Modeling and Optimization of an Ion Transport Membrane for Oxygen Separation from Air

Dennis Loevlie

AIChE National Conference, 2018

Poster

Projects

Gaussian Splatting PyTorch

Implementation of 3D Gaussian Splatting for real-time radiance field rendering in PyTorch.

PyTorch 3D Vision Rendering
Code

Depth Anything Video

Monocular depth estimation for video sequences using the Depth Anything model.

Computer Vision Depth Estimation Video
Code Demo

GPT4Readability

CLI tool leveraging LLMs and vector databases with LangChain and llama.cpp to generate README files and suggest code improvements. Supports cloud and local open-source LLMs across 15 languages.

NLP LangChain Open Source
Code Demo

SkinsAI

Free diagnosis tool classifying moles as benign or malignant using a PyTorch CNN. Won 2nd place out of 24 teams at The Pitt Challenge Hackathon.

Medical AI PyTorch Hackathon
Code Competition Devpost

Aqueous Solubility Prediction

Machine learning models for predicting aqueous solubility of chemical compounds.

Cheminformatics ML Chemistry
Code Kaggle

ASL Active Learning

Active learning framework for American Sign Language recognition with efficient labeling.

Active Learning Computer Vision ASL
Code

Disparity Maps

Stereo vision disparity map generation for depth perception from image pairs.

Computer Vision Stereo Depth
Code Course PDF

QA Model

Question answering system built with transformer-based architectures.

NLP Transformers QA
Code Course

Face Verification

Face verification system using deep metric learning and siamese networks.

Computer Vision Biometrics Deep Learning
Code Course

Rhine Water Level Prediction

Time series forecasting for Rhine river water levels using ML models.

Time Series Forecasting Environmental
Code

Speech Recognition

Automatic speech recognition system using deep learning architectures.

Speech Deep Learning Audio
Code Course

Speech to Text

End-to-end speech-to-text transcription pipeline with attention mechanisms.

Speech NLP Seq2Seq
Code Course

ML Codes

Collection of machine learning algorithm implementations from scratch.

ML Algorithms Education
Code Course

Patient Stay Prediction

Predicting patient length of stay in hospitals using clinical data and ML.

Healthcare ML Prediction
Code Slides

nb_search

Search tool for finding content across Jupyter notebooks in a repository.

Developer Tools Jupyter Search
Code PyPI

Interactive Demos

AuPd Nanoparticle Viewer

55-atom Mackay icosahedron — the structure studied in Demystifying the Chemical Ordering of Multimetallic Nanoparticles. Drag to rotate, scroll to zoom.

CE = ... eV/atom
Au Pd

Depth Estimation

Client-side monocular depth using Depth Anything V2. Model loads on first use (~27 MB).

Drop an image or click to upload

Experience

2026 —

ELLIS PhD Student

CWI & University of Amsterdam

Researching multimodal tabular foundation models. Advised by Madelon Hulsebos (TRLab, CWI) and co-advised by Jan-Willem van de Meent (AMLab, UvA).

2025

Multimodal AI Intern

ContentsPal

MIT professor-led AI startup in the insurance space. Implemented learning-based duplicate detection and open-vocabulary instance segmentation.

2024 —

M.S. Computer Science

Tufts University

GPA: 3.95. Research on deep MIL for computational pathology with Dr. Michael Hughes. Also working with Dr. Jivko Sinapov on vision-language models and reinforcement learning.

2023 — 2024

Senior CV & ML Engineer

KEF Robotics

Led a team of five engineers on a $500K NAMC project. Developed on-device object detection, monocular depth prediction, and 3D map generation for autonomous UAVs.

2021 — 2023

Graduate Research Assistant

University of Pittsburgh — CANELa Lab

Applied ML, Boltzmann statistics, and evolutionary optimization to predict material properties of metal nanoparticles with Dr. Giannis Mpourmpakis.

2020 — 2021

Lead Data Scientist

AiThElite

Pittsburgh-based startup using AI to improve the college athlete transfer process. Built the frontend and backend with Django, hosted on AWS.

2019 — 2020

M.S. Chemical Engineering

Carnegie Mellon University

GPA: 3.91. Research with Dr. John Kitchin on software tools for high-throughput experiments. Developed nb_search Python package.

2017 — 2019

Undergraduate Research

West Virginia University

Mathematical modeling and optimization with Dr. Fernando Lima. Won 2nd place at AIChE National Poster Competition.

2016 — 2019

B.S. Chemical Engineering

West Virginia University

Graduated with Honors, Cum Laude.

Open Source

TorchMIL

Contributor

A PyTorch library for deep multiple instance learning in computational pathology. Includes standardized benchmarks and reproducible implementations of attention-based MIL methods.

HuggingFace Transformers

Contributor

Contributed to the HuggingFace Transformers library, the leading open-source platform for state-of-the-art NLP models.

Get in Touch

Whether you're interested in collaborating on research, have questions about my work, or just want to say hello — I'd love to hear from you!

[email protected]