# Christopher P.

> Software Engineer II at Hive

Location: San Francisco Bay Area, United States
Profile: https://flows.cv/christopherp2

Undergraduate student at UC Berkeley interested in computer science, engineering, and artificial intelligence.

## Work Experience
### Software Engineer II @ Hive
Jan 2024 – Present | San Francisco, California, United States

### Software Engineer @ Hive
Jan 2023 – Jan 2024 | San Francisco, California, United States
ML Infra/Vector Search

### Software Engineer Intern @ Roblox
Jan 2022 – Jan 2022
• Created a command-line interface for utilizing internal tool with NodeJS, which made using the tool significantly easier and reduced onboarding time from multiple days to minutes
• Improved and fixed various parts of Roblox’s thumbnail generation pipeline with C# and JavaScript such as thumbnail invalidation and alert generation

### Software Engineer Intern @ Apple
Jan 2021 – Jan 2021
• Created VSCode extension for assisting data scientists working in internal Clojure domain specific language using TypeScript
• Improved productivity and reduced errors by adding code snippets, autocomplete, and error checking features
• Lessened friction in testing production by implementing one-click end-to-end operations

### Software Engineer Intern @ Apple
Jan 2020 – Jan 2020
• Migrated ETL service to cloud using AWS and Kubernetes to improve performance
• Setup CI/CD pipeline for service deployed with Kubernetes/AWS Elastic Kubernetes Service and
improved event loop with Java
• Collaborated with other interns to create microservice for internal team that retrieves,
crop/rotates, and scores images using Python

### Research Intern @ Harvard T.H. Chan School of Public Health
Jan 2018 – Jan 2018
• Participated in a healthcare research project utilizing Natural Language Processing techniques to classify large volume corpora of patient files into categories of diseases.
• Designed and implemented 5000+ lines of code in Python and R with 500+ hours of research.
• Experimented with regression models and a novel approach to boost traditional Latent Dirichlet Allocation (LDA) algorithm performance by using expert knowledge and prior probability on relevant concepts. Prior probability was trained from public domain data including Medline, Medscape, Merck Manual and Wikipedia. Large scale pre-processing of 50,000+ medical articles representing 1.5MM+ Concept Unique Identifiers was used to identify concepts that best describe each disease. Constructed inverse probability training samples based on correlation to main disease. Fifty-fold cross validation with Elastic LASSO augmented with hyperparameter optimization was used to select relevant concepts and train weights.


## Education
### Bachelor of Science - BS in Electrical Engineering and Computer Science
University of California, Berkeley

### High School
Mission San Jose High School


## Contact & Social
- LinkedIn: https://linkedin.com/in/chris-pan79
- GitHub: https://github.com/chris-pan

---
Source: https://flows.cv/christopherp2
JSON Resume: https://flows.cv/christopherp2/resume.json
Last updated: 2026-04-11