About
Who I am
Computer Science student at UC Irvine focused on building scalable software systems that solve real problems. I enjoy working across AI, backend infrastructure, and full-stack development to turn complex ideas into clean, reliable products.
Outside of tech, I enjoy traveling, working out, listening to music, playing video games, and watching Formula 1 and cricket.
San Francisco Bay Area, CA
viditjain2005@gmail.com
⌘
3+
Years Coding
◈
3
Research Roles
◻
10+
Projects Built
◉
3.7
GPA
Experience
Where I've worked
Software Engineer
Jan 2025 – May 2026
UC Irvine — Spider Lab
- Built a fault-tolerant GitHub GraphQL ingestion pipeline processing 200K+ commits across 100K+ repositories, with token rotation and checkpoint recovery improving pipeline reliability by 4×
- Engineered large-scale AST-based static analysis pipelines across multi-language repositories, applying semantic filtering to reduce false positives by 30%
- Optimized repository mining pipelines with streamed processing, automated failure recovery, and resumable checkpointing
AI Engineer (Research Assistant)
Oct 2025 – Dec 2025
UC Irvine
- Built Python pipelines to analyze 2,500+ unmerged PRs across agents to identify systemic CI/CD failure patterns
- Developed classification workflows identifying 31% Cursor merge conflicts and 16% Claude test failures
- Automated mining of 1,300+ maintainer comments, revealing 38% of PRs were rejected without reasoning
Software Engineer (Research Assistant)
Feb 2025 – May 2025
UC Irvine — UCIPT
- Fine-tuned transformer models for tone and sentiment detection, improving classification accuracy on dialogue data
- Built speech-to-text pipelines with emotion tagging, enabling contextual understanding for voice-driven AI systems
- Integrated NLP models into real-time chatbot workflows, improving response relevance and conversational coherence
Projects
Things I've built
IN PROGRESS
PitIQ
An F1 race strategy engine that uses multi-agent simulation and reinforcement learning to recommend optimal pit stop strategies tailored to individual driver styles.
PythonFastAPIXGBoostReactTypeScriptFastF1
TrueShuffle
Built a cross-platform desktop app that enhances Spotify shuffle using smarter queue balancing to reduce repetition and improve listening variety.
TypeScriptReactTauriRust
Waypoint
An AI-powered travel planning platform that generates personalized itineraries from user preferences, travel dates, and location context in real time.
PythonReact NativeExpoNode.js
Reddit Guard
A real-time toxicity analysis tool that detects harmful language in Reddit communities using transformer-based NLP models, achieving 90% accuracy.
PythonTransformersPRAWStreamlitNLP
Parq
A minimal iOS parking assistant that helps users remember where they parked, track meter time, save garage photos, and receive smart leave-now alerts.
SwiftMapKitStoreKit
XCleaner
AI-powered file organizer that automatically sorts documents and media into custom categories using NLP, machine learning, and content analysis.
PythonNLP
Skills
Tech stack
⌨Languages
PythonJavaC/C++SQLJavaScriptTypeScriptR
◻Frontend
ReactReact NativeNext.jsExpoNode.jsTailwind CSS
◈Backend
FastAPIFlaskNode.jsExpressREST APIs
◉AI / ML
PyTorchPandasNumPyTransformersTensorFlow
◫Infrastructure
DockerAWSCI/CDGit
◧Tools
CursorClaude CodeExcel