Vin̆it Üpadhy̆åy̆
Creating with code. Small details matter.
Overview
Social Links
About
Hello, World! I am Vinit Upadhyay — a Front-End Developer passionate about creating high-performance, user-centric software solutions with intuitive and engaging designs.
Currently working as an AI Trainer & Reviewer at Outlier, where I collaborate with top AI organizations to enhance LLM through expert-level front-end development insights. I contribute to the training of generative AI by creating and answering questions related to web development, evaluating AI-generated code, and reviewing UI/UX quality.
Previously, I worked as an SDE (Frontend Engineer) at SameSpace, where I engineered UI components for the Wave analytics platform, resulting in 20% faster page loads and 15% higher user engagement. I optimized frontend with Tanstack Query & GraphQL, reducing API call overhead by 35% and data transfer by 40%.
I specialize in modern front-end technologies including React.js, Next.js, TypeScript, and Tailwind CSS, with a strong focus on responsive design, accessibility (WCAG-compliant), and best UX practices. I'm also proficient in backend technologies like Firebase, Express.js, and databases like MySQL and MongoDB.
One of my key projects, QuickTube, is a platform streamlining video review and approval between YouTubers and editors, reducing the review process time by 40%. It integrates seamless YouTube channel connectivity, editor access to workspaces, and a notification system via WhatsApp or Slack.
I'm currently pursuing a BS in Data Science and Application at IIT Madras, and hold a BSc in Computer Science from SIES College of Arts, Science, and Commerce.
Let's connect and collaborate!
Stack
Experience
Outlier
Current Employer- Leveraged GenAI and LLM-based frameworks to evaluate AI-generated code, prompting models to reason, refactor, and improve solutions across real-world engineering tasks.
- Spearheaded the development of novel prompt engineering techniques, enhancing model reasoning accuracy by 20% on complex code generation tasks, improving overall code quality and model usefulness.
- Collaborated with senior research teams and domain experts, aligning evaluation criteria with industry-quality benchmarks to advance enterprise-grade AI models.
- React
- TypeScript
- Next.js
- HTML
- CSS
- JavaScript
- UI/UX Design
- Accessibility
- AI Training
- Code Review
SameSpace
- Engineered UI components for Wave analytics platform, resulting in 20% faster page loads and 15% higher user engagement.
- Optimized frontend with Tanstack Query & GraphQL, reducing API call overhead by 35% and data transfer by 40%.
- Developed 15+ reusable, accessible (WCAG-compliant) UI components, accelerating delivery across 8 product modules by 25%.
- Monitored frontend performance using browser dev tools, Lighthouse, and custom metrics logging; improved time-to-interactive (TTI) by 22%.
- React
- TypeScript
- Tanstack Query
- GraphQL
- WCAG
- Performance Optimization
- Lighthouse
- UI Components
- Analytics
Education
- Currently studying for a BS in Data Science and Application.
- Located in Chennai, India.
- Data Science
- Statistics
- Machine Learning
- Python
- R
- SQL
Personal Projects(2)
- Architected full-stack e-commerce solution using Next.js 15, React 19, and TypeScript, implementing object-oriented design patterns and component-based architecture for maintainable, scalable codebase.
- Integrated GraphQL API with Prisma ORM and PostgreSQL database, achieving 40% faster data retrieval compared to traditional REST approaches while reducing query complexity by 35% through optimized schema design.
- Devised and executed a comprehensive testing strategy including unit tests with Jest and integration tests, ensuring 85%+ code coverage and maintaining software quality standards throughout development lifecycle.
- Next.js 15
- React 19
- TypeScript
- Tailwind CSS
- Framer Motion
- GraphQL
- Prisma ORM
- PostgreSQL
- Supabase
- Row Level Security
- React Query
- Docker
- WCAG
- Spearheaded a team to develop QuickTube, a platform streamlining video review and approval between YouTubers and editors, reducing the review process time by 40%.
- Integrated seamless YouTube channel connectivity, editor access to workspaces, and a notification system via WhatsApp or Slack, improving collaboration efficiency by 30%.
- Facilitated video metadata customization, a two-step approval process, and enhanced security for content before YouTube upload, reducing content errors by 20%.
- Leveraged React.js, Node.js, Express, Firebase, YouTube API, Twilio API, and Slack API in the technology stack, ensuring scalability and performance.
- React.js
- Node.js
- Express
- Firebase
- YouTube API
- Twilio API
- Slack API
- Real-time Collaboration
- Video Review
Client Projects(4)
- Built a professional project estimation tool with Next.js 15 (React 19, TypeScript) featuring feature toggles and per-feature complexity sliders.
- Generated language-specific pseudocode (TypeScript, Python, Java, Go, Ruby) using Prism.js syntax highlighting.
- Calculated methodology-based time estimates (Waterfall, Agile, Kanban, XP) with JSON export for client-ready deliverables.
- Implemented responsive UI with Tailwind CSS, draggable split pane, and accessible controls.
- Next.js 15
- React 19
- TypeScript
- Tailwind CSS
- Prism.js
- Responsive UI
Certifications(4)
Vin̆it Üpadhy̆åy̆
Frontend Engineer