Amogh
A fourth year at University of Virginia passionate about building software and using technology to solve problems. Look around to see what I’ve been working on!
Hi, I'm
Education
Aug 2022 - May 2026
University of Virginia
B.S. in Computer Science & Minor in Data Science
GPA:
3.98 out of 4.00
Relevant coursework:
- Data Structures & Algorithms 2, Advanced Algorithms, Advanced Software Development, Computer Architecture & Systems 2, Discrete Mathematics 2, Multivariable Calculus, Statistics, Linear Algebra, Machine Learning, Artificial Intelligence, Databases, Programming Languages, Web Applications, Cloud Computing, Reinforcement Learning, Natural Language Processing, Parallel Computing, Compilers
Experience
- Building a mobile app feature for over 35,000 daily users to book study spaces in Capital One Cafés through a conversational experience, implemented natively in Swift and powered by an AI agentic backend
- Fine-tuned LLaMA 3 with Quantized Low Rank Adaptation (QLoRA) on company emails to build an internal tool that generates sentiment-aware, on-brand email replies, improving client response time by 45%
- Improved effectiveness of AI advertisement generator by switching to multi-agent workflows with LangGraph, LLaMA 3, and Chroma DB; used to create 13 real ads reaching over 50,000 impressions across 4 client campaigns
- Deployed Generative AI chat systems for clients to provide accurate responses about their reports and blog posts using retrieval techniques with LangChain, Pinecone, and GPT-4; serving a total of over 20,000 users
- Created an advertisement generator web app powered by GPT-4 and DALL-E APIs, used to generate more than 100 concepts for the advertising campaigns of over 20 different customers
- Developed news feed using Twitter’s v2 API on ABC Action’s mobile app for the association’s 22,000 members
- Added insightful statistics, recommendations, and design changes to the platform for increased savings engagement, as measured by 650 new employee subscriptions and a 15% increase in users saving towards two or more goals
- Fixed critical TypeScript logic errors and Angular/Bootstrap design bugs leading to authentication issues
Projects
FundHer
Platform for students to easily find personalized scholarships, grants, and other funding opportunities. Algorithm connects users to opportunities best matching profile/background with 93% accuracy via vector search.
Next.js
Tailwind CSS
TypeScript
FastAPI
MongoDB
Pinecone
Node.js
Puppeteer
Messaging Platform
A real-time messaging app where users can sign in with Google and create group chats or direct messages with other users.
React.js
TypeScript
GraphQL
Python
AWS
Cognito
AppSync
Lambda
DynamoDB
Spotify Music Recommender
A machine learning content-based recommendation system using k-means clustering and cosine similarity. Invoked from web app that uses Spotify API to allow users to link a playlist and get new song recommendations based on the playlist.
React.js
Spotify API
Scikit-Learn
AWS
SageMaker
API Gateway
Lambda
IntelliGrade
A gradebook for teachers to create assignments and for students to upload their work. Automatically grades assignments using a vision model to transcribe work and an LLM to evaluate it against a rubric along with feedback.
Next.js
TypeScript
Tailwind CSS
Python
LangChain
AWS
S3
API Gateway
Lambda
DynamoDB
Market Performance Visualizer
Data aggregation tool that graphs economic indicators and stocks, letting users analyze market performance. Makes API calls to the Federal Reserve Economic Data (FRED) for most up-to-date statistics and stock data.
React.js
Recharts
FRED API
AWS
Route 53
CloudFront
S3
Awards & Certifications
Get in Touch
Feel free to reach out to me!