Software Engineer · USC MSCS · 2026
Vaidik
Vadhavana
I build scalable software that ships and lasts.
Currently Building
About
I'm a Master's student in Computer Science at USC (graduating May 2026), building at the intersection of backend systems, developer tooling, and applied AI. I've shipped VS Code extensions to the Marketplace with 350+ installs, built full-stack platforms for hundreds of users, and conducted ML research at USC Viterbi achieving 98.54% diagnostic accuracy.
Currently at Exelint International, I've connected Oracle NetSuite to Claude AI via Model Context Protocol, shipped a RAG chatbot for e-commerce, and automated workflows that eliminated weeks of manual accounting cycles. When I'm not coding, I shoot film photography, explore music, and find trails.
Focus Areas
- Backend Systems
- Developer Tooling
- AI / LLM Integration
- Applied ML Research
Selected Work · 2023–2026
DEVELOPER TOOLING
OmniChat AI
350+ installs · 6 GitHub stars
A privacy-first VS Code extension enabling offline AI assistance by integrating any local LLM through Ollama — including OpenAI's open-weight models. Built for engineers who want AI without the cloud.
BROWSER EXTENSION
ZenMark
20+ active users · <100KB build
Chrome extension with a custom Trie-based fuzzy search engine delivering <15ms latency across 10,000+ bookmarks. Privacy-first, fully local, no external APIs.
PRODUCTIVITY · WINDOWS
EzTranslate
10+ languages · fully offline
Native Windows utility that auto-detects clipboard content and returns instant translations via a fine-tuned Gemma3 model. No internet, no API calls.
FULL-STACK PLATFORM
APSIT Community
200+ active student users · AWS deployment
Full-stack social platform enabling inter-department communication at university. JWT-secured, deployed on AWS with Amplify + Elastic Beanstalk. Led a team of 4 engineers.
ML · MOBILE
NutriLytik
170+ patients surveyed · 10 beta testers
Mobile app enabling diabetic patients to scan food labels for real-time nutritional insights via OCR. Personalized meal plans generated from ML models trained on 170+ patient surveys.
OSS Contributor · Omi AI Ecosystem (2 apps shipped) · Invited to SF Office · May 2025
Timeline
Education
M.S. Computer Science
University of Southern California
B.E. Computer Engineering
A. P. Shah Institute of TechnologyGPA 9.28 / 10
Experience
Software Engineer Intern (Co-op)
CURRENTExelint International
Pioneered NetSuite × Claude AI integration via Model Context Protocol (MCP) with OAuth 2.0 and custom SuiteApps. Shipped a production RAG chatbot for e-commerce support query resolution. Eliminated a 15–18 day manual accounting cycle through native NetSuite automation — projected to close books in under 4 days.
Research Assistant
USC Viterbi School of Engineering
Built a hybrid deep learning model for early-stage glaucoma detection in retinal fundus images — 98.54% accuracy. Extended to Diabetic Retinopathy detection by merging 3 datasets to resolve class imbalance, improving accuracy by ~5%.
Full-Stack Developer (Project Lead)
A. P. Shah Institute of Technology
Led a team of 4 to build and ship APSIT Community — a full-stack social platform on AWS growing to 200+ active student users.
Technical Depth
Languages
- Python
- TypeScript / JavaScript
- Go
- C / C++
- C#
- SQL
Infrastructure
- AWS (EC2, S3, Lambda)
- Docker · Kubernetes
- Terraform
- NetSuite REST API
- CI/CD (GitHub Actions)
- MongoDB · PostgreSQL · Redis
AI / ML
- PyTorch · TensorFlow
- LangChain · RAG Pipelines
- Ollama (local LLMs)
- Model Context Protocol (MCP)
- OpenCV
- Hugging Face
Contact
Let's build something.
Always open to interesting conversations,
collaborations, and ideas.