Harshad Selokar
Crafting innovative solutions at the intersection of technology and creativity. Passionate about building scalable applications and exploring the frontiers of AI.
About Me
I'm a Full Stack Developer & AI Engineer focused on building practical products with React, Node.js, FastAPI, and modern AI workflows. My work combines web development, agentic AI, and RAG-based systems to deliver scalable and useful user experiences.
I have built production-ready solutions such as a college portal with secure authentication, SQL-backed workflows, and file uploads, along with AI-powered projects like adaptive RAG chatbots and content-driven knowledge tools. I also contribute to chapter websites, technical workshops, and performance-focused web interfaces.
Focus Areas
- • Full-Stack Web Development
- • Agentic AI & RAG Systems
- • Backend Engineering
Interests
- • System Design
- • DevOps
- • AI & Automation
Highlights
- 6x hackathon winning and 27 participations
- Open Source Contributor
- AI & RAG Developer
- Full Stack Engineer
Experience
Software Developer Intern
TPDC SVPCET Nagpur
Built a production-grade college portal using Node.js, Express, and EJS, streamlining gatepass, attendance, room booking, and complaint workflows. Implemented secure authentication using bcrypt and session-based middleware. Designed SQL-based relational database with schema and migrations. Developed REST-style endpoints supporting 100+ daily user requests with file upload functionality.
Web Master & Treasurer
ACM Association for Computing Machinery, SVPCET
Architected and deployed ACM chapter website using React.js and Node.js, increasing member engagement by 40% and supporting 100+ active users. Implemented CI/CD pipeline using GitHub Actions, reducing deployment time by 60%. Managed backend with Express.js and MongoDB handling 100+ monthly API requests at 99.9% uptime; achieved 95+ Lighthouse score. Conducted 5+ technical workshops on Prompt Engineering and Competitive Coding, training 200+ students.
Project Intern - Software Developer
Vectortek Labs Pvt Ltd
Developed frontend modules using React.js and TypeScript, improving internal workflow efficiency by 35% for 50+ team members. Integrated REST APIs with Node.js backend, reducing data fetch latency by 45% through efficient state management and caching. Wrote unit tests using Jest and React Testing Library achieving 85% code coverage and reducing production bugs by 30%.
Frontend Developer
Unnat Bharat Abhiyan, SVPCET
Engineered responsive government initiative website using React.js and TailwindCSS, achieving a 98% accessibility score and LCP < 2.5s. Implemented SEO best practices resulting in 150% increase in organic traffic within first 3 months.
Website Advisory & Developer
TECHNEX SVPCET (Annual Technical Fest)
Built 15+ responsive web pages using React.js and TailwindCSS for 2,000+ concurrent event users with zero downtime. Implemented advanced animations using GSAP and Three.js, increasing user engagement by 55%. Optimized bundle size by 40% through code splitting and tree shaking, improving load time from 4.2s to 2.5s.
Web Development Intern
Bharat Intern
Completed frontend development projects using HTML5, CSS3, and JavaScript with focus on responsive design, accessibility, and cross-browser compatibility.
Skills & Technologies
HTML5
CSS3
JavaScript
TypeScript
React
React Native
Redux
Bootstrap
TailwindCSS
GSAP
Three.js
Express.js
Node.js
FastAPI
MongoDB
Firebase
Git
GitHub
Vercel
Remix.js
Python
C
C++
Arduino
Microsoft Excel
Power BI
SQL
PostgreSQL
EJS
GitHub Actions
Featured Projects
FinPilot AI - Personal Finance Management & Advisory System
AI-based personal finance system for tracking expenses, managing budgets, detecting risks, and improving financial decision-making.
SiteSage
AI knowledge assistant that unifies websites, YouTube, and PDFs with RAG. Built with FastAPI, Next.js, Supabase, Groq, and a Chrome extension for smart snapshots.
CoExist - Reducing Human-Wildlife conflict.
AI and IoT wildlife monitoring platform that tracks animal movement in real time, predicts conflicts, and sends instant alerts to villagers and forest officials.
Adaptive RAG - Agentic AI Chatbot
Adaptive RAG chatbot built with FastAPI, LangGraph, LangChain, Qdrant, MongoDB, and Streamlit for retrieval, web search, and persistent chat memory.
Prompt Spark Engine
Prompt Spark Engine is a tool for testing and optimizing prompts using local LLMs like Mistral, with real-time scoring and visualization.
Gate Pass System
Gate Pass System is a Flask-based app using QR codes to track student entry and exit, with auto check-in/out, QR generation, and 3-day activity history.
Get In Touch
Let's Connect
I'm always interested in discussing new opportunities, innovative projects, and potential collaborations in AI and software development.