Projects

ALAN: Academic Learning Analysis Network

Jan. 2026 - May 2026

Skills used: Python, Bittensor, Hugging Face, Docker, FastAPI, Git

Developed a decentralized Bittensor subnet implementing miner-validator evaluation to incentivize high-quality academic journal peer reviews
Deployed and tested a local Subtensor blockchain with Docker, integrating wallet management, subnet registration, and metagraph interactions
Engineered LLM-powered validator pipelines to score peer reviews using multi-criteria evaluation and on-chain reward distribution
Built data ingestion pipelines using the OpenReview API to preprocess academic review datasets for model evaluation

Oct. 2024 - April 2025

Skills used: JavaScript, AWS Lambda, API Gateway, DynamoDB, AWS S3, React.js

Developed a full-stack serverless web app using AWS Lambda and API Gateway for backend and DynamoDB for data storage, enabling users to share and explore geotagged locations.
Created an interactive map frontend with Leaflet.js by creating custom React components for map rendering and user interactions.
Accomplished geospatial data storage using DynamoDB, indexing with Leaflet.js, and AWS S3 for image storage.

Jun. 2024 - Aug. 2024

Skills used: Python, AWS Lambda, AWS API Gateway, Canvas REST API

Designed a voice-enabled Echo skill for accessing student Canvas schedules using Python and AWS Lambda.
Utilized Canvas’ LMS REST API for immediate schedule information retrieval.
Enabled serverless processing using AWS Lambda and handled Canvas API requests with API Gateway.
Implemented secure login to manage user permissions and protect sensitive academic data.

Apr. 2024

Skills used: Python, Git, Tkinter, Pandas

Built a Python application with a Tkinter GUI allowing users to input ingredients and search a dataset of 100k+ recipes, discovering recipes with the highest number of matching ingredients.
Used graphs, maps, and priority queues to optimize the search algorithm and improve performance.
Utilized Pandas for efficient data manipulation and processing of large datasets, enabling rapid search responses.
Achieved 3rd place overall in a competition out of 200+ students in a Data Structures & Algorithms class.