Tim Wang

My interests lie in understanding, visualizing, and processing large datasets quickly, efficiently, and comprehensively. I currently work on the Bioinformatics and Computational Biology team at DEM BioPharma, where we work to derive insights from high-throughput RNA sequencing (RNA-Seq) and CRISPR screening data from multi-armed experiments.

Previously, I worked at the Broad Institute of Harvard and MIT Data Sciences Platform as a software engineer, helping with the day to day testing of improvements to production-grade high-throughput sequencing data processing pipelines.

2024/01/26

This site was made for a data visualization course! Things that might be fun to add later:

Tacky animations
Embedded things like GitHub commits or LinkedIn posts
Link to public calendar whenever that gets made

Bioinformatics & Computational Biology Co-op, DEM Biopharma

Containerize custom bioinformatics analysis tools to facilitate pipeline migration to AWS Batch
Perform statistical validation of quality control metrics for newly migrated pipelines to ensure accuracy/li>
Design static and interactive visualization notebooks for downstream analysis of pipeline outputs
Create ad-hoc data processing and analysis scripts for speeding up menial tasks for other teams

Software Engineering Co-op, Broad Institute of Harvard and MIT

Maintained scalable analysis pipelines for processing high-throughput sequencing data
Standardized Docker images and infrastructure for scientific tools to align with best practices
Monitored and addressed periodic security vulnerability (CVE) scans in Docker images
Productionized Jupyter notebook analysis steps into reusable command line scripts

Classifying Acute Leukemia, on GitHub

Used pre-normalized gene expression data to predict Acute Myeloid/Lymphocytic Leukemia
Compared logistic regression, random forest, KNN models, and a dense neural network for accuracy
Achieved 97\% accuracy with the logistic regression model, trained on only 72 data points

Bootstrap Percolation Algorithms, an NSF-funded REU

Implemented a basic percolation process in Python for large, randomly generated graphs
Conducted empirical investigation of algorithms presented by leading papers in the field
Developed a custom percolation algorithm, improving performance on certain classes of graphs

Predicting Logan Airport Delays, on GitHub

Investigated relationships between weather conditions and arrival delays at Boston Logan Airport
Visualized feature/outcome variable relationships in US Department of Transportation data
Optimized and compared Ridge, Lasso, and Linear regression algorithms for delay prediction

Transit Ticketing Database, on GitHub

Designed a relational database schema to model a public transit fare system
Integrated data model with a CRUD-operable backend and React-based web interface

In-progress

Advanced Linear Algebra
Algorithms and Data
Data Visualization and Information Presentation
Practical Neural Networks

Completed (not exhaustive)

Machine Learning and Data Mining 1
Large Scale Storage and Retrieval
Database Design
Object-Oriented Design
Advanced Group Theory
Complex Variables
Statistics & Stochastic Processes
Dynamical Systems
Real Analysis