I'm Gabriel. I'm a Software Engineer in NYC. Previously, I graduated in May 2025 from Columbia University with a B.S. in Computer Science.
- Incoming Software Engineer @ Amazon (July 2025, New York)
- Software Engineer Intern @ Amazon (Summer 2024, New York)
- Teaching Assistant @ Columbia (multiple semesters, several Physics and Computer Science classes)
-
culpa.info (>30k monthly users)
Columbia’s course review platform, rebuilt with a modern stack and AWS infrastructure. The respository for this project is private, but you can check out this blog post that describes the system design.
My contributions:
- Designed and implemented all cloud infrastructure with 100% IaC (AWS CDK). Used CloudFront, EC2, API Gateway, RDS, S3, and CloudWatch. See details in this blog post.
- Added caching to our backend for common classes and professors. This significantly decreased load in our database and improved peak traffic performance a lot.
- Wrote Github Actions scripts to make testing/deployments automatic. Decreased deployment time from ~1 hour (manual deployments) to < 5 mins (automated deployments).
- Implemented Professor overview feature, which generates AI overviews for each professor from their reviews with Gemini API. Check out this blog post.
-
sbsim is Google open-source lightweight physics-informed simulator used to train reinforcement learning algorithms to control HVAC system.
My contributions: I added the reinforcement learning module, which is the code used to train and evaluate RL agents. Also contributed Github Actions CI/CD scripts for automated testing, other qol improvements. Multiple merged PRs.
-
The documentation website for the sbsim project.
My contributions: the documentation for the reinforcement learning module.
-
Unofficial Python client library for Semantic Scholar APIs.
My contributions: added support for async calls. Multiple commits/PRs merged.
