Data Science and ML
Recent Caltech grad
Formerly at Recursion Pharmaceuticals
Data Scientist and recent Caltech graduate with five years of experience building data and machine learning solutions in academic and industry settings. Skilled at developing pipelines for processing and analyzing large datasets across biological, geospatial, and astrophysics applications. Experienced in applying advanced machine learning techniques, including state-of-the-art vision and language models. Co-founder of a pre-seed AI startup, I designed and engineered our machine learning models, data pipelines, databases, and underlying infrastructure. Significant experience developing statistical models to extract insights from complex datasets, including teaching graduate students at Caltech. With a diverse set of experiences, a deep skill set, and unmatched determination, I am equipped to tackle any challenge that I face and make a major impact on any team.
During my time at Caltech, I had a chance to experience roles outside the classroom in both academia and industry. My diverse experiences have given me a strong skillset and background to draw from, enabling me to rapidly integrate and contribute value to any team.
September 2023 - December 2023
• Instructed 85 students in the graduate-level course Introduction
to Data Analysis in
the Biological Sciences, developing their skills in statistical modeling, numerical
optimization, data visualization, and exploratory data analysis in Python
June 2023 - September 2023
• Developed 2 new statistical metrics for analysis of large
experimental datasets that were used to improve model performance
• Implemented advanced machine learning techniques in PyTorch to
improve data processing efficiency and reduce costs
• Deployed monitoring tools to the data science and QA teams in
collaboration with a 20-person cross-functional team, ensuring data integrity for
downstream processing and models
May 2022 - September 2022
• Built data pipeline leveraging deep learning and statistical
modeling to accelerate image processing speed by a factor of 100
• Automated microscopy analysis and increased measurement
accuracy by an estimated 10% for an upcoming publication
May 2020 - January 2021
• Implemented statistical analysis methods using common data science
tools including Python, NumPy, pandas, SciPy, and Jupyter notebooks on HPC clusters
to perform source classification and deliver insights from large datasets
• Initiated development of an end-to-end data pipeline for
automated data processing and classification, improving processing speed and
efficiency in handling large datasets and decision-making
With a wide variety of experiences to draw from, I have worked extensively with the core data science stack while also being exposed to a many of the technologies used across industry.
September 2023 - December 2023
• Co-founder and Machine Learning Engineer for pre-seed AI startup
for personalized second-hand fashion recommendations
• Transformed a business idea into reality by designing data
pipelines, ML training and inference frameworks, and PostgreSQL databases, all
custom-built to develop models using PyTorch on Google Cloud and serve them to the
production website
September 2023 - December 2023
• Led team of 3 in ML project to predict surface temperatures from
spectral data, achieving error of less than 1 C
• Created a new ML dataset from raw NASA data in collaboration
with JPL scientists to study of novel research questions
• Built model training and testing frameworks and designed
CNN-based model architectures alongside Professor Katie Bouman for use with
autoencoder-generated embeddings to deliver accurate predictions
March 2023 - June 2023
• Orchestrated training of multi-GPU PyTorch models on HPC clusters
using Slurm Bash scripts to enable powerful classifiers
• Developed a pipeline to process over 224k chest x-rays, adding
augmentations to further improve model diagnosis accuracy
• Placed 3rd as a solo contestant competing against teams of 5
in Caltech's annual machine learning competition