CV

Education

• MSc in Applied Computing (MScAC), Department of Computer Science, University of Toronto, 2023-2025
  • Cumulative GPA: 4.00/4.00
  • Relevant courses: Knowledge Representation and Reasoning, Automated Reasoning with Machine Learning, Neural Networks and Deep Learning, Software Engineering for ML-Enabled Systems, Visual and Mobile Computing
• BS in Data Science, Halıcıoğlu Data Science Institute (HDSI), University of California San Diego, 2019-2023
  • Cumulative GPA: 3.54/4.00

Publications

Preprints & In-submission

Active Learning for NANOBODY® Structure Prediction

Preprint - Submitted to Submitted to Research in Computational Biology (RECOMB)
[Preprint Link]
October 2024

[Under Review]

Preprint - Submitted to Submitted to International Conference on Learning Representations (ICLR)
[Preprint Link]
October 2024

Research & Work Experience

• Sanofi Canada, Toronto, Canada
  Computational Scientist (Co-op) May 2024 - December 2024
  • Developed DEWDROP, a novel active learning strategy that uses an ensemble of predictions to estimate epistemic uncertainty, optimizing in silico nanobody candidate development.
  • Outperformed BADGE, KMeans, and Random baseline strategies, reducing sample requirements by 25% for achieving target loss in nanobody structure prediction.
  • Fine-tuned the Nanofold model using DEWDROP, achieving a FAPE loss below 0.5 in just 5 rounds, utilizing 35% of the total dataset.
  • The DEWDROP-Nanofold pipeline is patented, and a manuscript is in preparation for submission to RECOMB.
• REAP Lab @ University of Toronto, Toronto, Canada
  Research Assistant October 2023 - May 2025
  • Developed Vi-SATNet, a neural-symbolic architecture, combining CNN models with SATNet to learn implicit propositional rules in the feature space.
  • Achieved over 90% accuracy on the MNIST dataset using feature regeneration, even when 80% of the features were masked.
  • Work submitted to ICLR 2025 for review.
• Abarbanel Computational Biophysics Lab @ UCSD, San Diego, CA
  Research Assistant September 2022 - June 2023
  • Developed a fast machine learning model based on the RBF network to simulate songbirds’ singing behavior.
  • Implemented GPU acceleration in PyTorch, reducing experiment runtime by 60%.
  • Developed the False Nearest Neighbor implementation, speeding up hyperparameter search by 10x.
• Personalis, Menlo Park, United States
  Data Warehouse Engineer Intern June 2022 - August 2022
  • Developed ETL pipelines for processing gigabyte-scale human genome data between SQL servers and Google BigQuery.
  • Created a time series prediction model using Google Cloud’s Dataflow and PySpark ML, and a collaborative filtering model for recommending relational tables.

Skills

• Programming: Python, Java, SQL, C (basic)
• Frameworks/Tools: PyTorch, TensorFlow, Spark ML, Scikit-Learn, MMSeq2, ViennaRNA, BigQuery, Dataflow, Dagster, Selenium, AWS, Nvidia DGX

Research Projects

• GPT-STEVE1: Long Horizon Action through Prompting Pre-trained Agent in Minecraft
  • Injected GPT-4’s world knowledge through multi-modal prompts to enhance the long-horizon action capability of STEVE-1 in Minecraft.
  • Demonstrated GPT-4’s ability to compose feasible action plans for complex tasks.
  • Identified limitations in VPT pre-training for action generalization.
• DeepCache-MLX: Accelerated Stable Diffusion on Apple Silicon
  • Developed DeepCache, a lightweight framework in MLX that caches high-level features during denoising, resulting in a 10x speedup in Stable Diffusion on Apple Silicon devices.
• Discovering Continuous Latent Space Representations for Graphs
  • Integrated autoregressive graph generation with Wasserstein GAN and VAE to benchmark model expressivity using MUTAG, ENZYME, and PROTEIN datasets.
  • Presented results at HDSI Capstone Showcase.
• Automate Image Captioning on Flask-based Web Application
  • Enhanced the open-source Albumy platform with BLIP-2 image captioning and search functionality using DETR for object detection.

Service and Leadership

• Toronto AI Practitioner Network (TAPNET), Toronto, Canada
  Organizer September 2024 - Present
  • Organized events connecting AI practitioners (academics, engineers, and founders) in Toronto and across Canada, fostering collaboration between professors, students, and the AI community.