I'm a final-year Information Technology undergraduate at Don Bosco Institute of Technology, Mumbai (CGPI 9.06/10.0), specializing in full-stack web development, applied AI/ML, and game development. My engineering focus centers on shipping scalable, production-grade systems β from multi-vendor marketplaces on Next.js 15 to RAG-powered chatbots serving government scheme data to real-time hospital dashboards.
I approach engineering with a product mindset: every system I build is evaluated not just on whether it works, but on latency, security posture, and the actual problem it solves for the end user. I'm a 3x hackathon finalist, an active open-source contributor, and I've spent the last few months building AI-moderated, real-time systems in a live production CRM.
π― Open To:
- Software Development Engineer (New Grad / Intern) roles
- AI/ML & Applied GenAI Engineering roles
- Open-source collaboration on full-stack / RAG tooling
- Freelance full-stack product builds
| Domain | Proficiency | Details |
|---|---|---|
| RAG Systems | Advanced | Built a LangChain + Ollama + ChromaDB pipeline achieving 95%+ relevance on 500+ government scheme queries |
| LLM Agents / Agentic AI | Advanced | Completed intensive coursework on generative & agentic AI; ships agent-driven application logic |
| AI-Based Moderation | Intermediate | Designed an in-app chat moderation layer to flag suspicious conversations and prevent customer leakage in production |
| Computer Vision | Intermediate | YOLOv8 + OpenCV crowd-detection system built at a national hackathon (Top 6 of 40+ teams) |
π GoCart β Multi-Vendor E-Commerce Marketplace
Full-stack multi-vendor marketplace enabling independent sellers to manage inventory, orders, and analytics through isolated, role-based dashboards.
| Stack | Next.js 15, Prisma, Neon (PostgreSQL), Clerk, Stripe, Inngest, Redux Toolkit, Tailwind CSS v4 |
| Scale | Multi-tenant marketplace with independent seller storefronts |
| Performance | Optimized cross-device responsiveness via Tailwind v4 + Redux Toolkit state management |
| Security | Role-based access control (Clerk) isolating Shopper / Seller / Admin dashboards |
| Impact | Reliable webhook sync and order tracking via Stripe checkout + Inngest async jobs |
| Repository | github.com/kpcode11/GoCart |
Architected the platform end-to-end on Next.js 15 and Neon PostgreSQL, with Clerk driving strict role isolation across three distinct dashboard experiences, and Inngest handling background job orchestration so webhook-driven order state never drifts from Stripe's source of truth.
πΎ Farmer Sahayak β Government Scheme Discovery Platform
An AI-powered discovery platform helping users find and understand relevant government welfare schemes through a retrieval-augmented chatbot.
| Stack | MERN Stack, Python Flask, LangChain, Ollama, ChromaDB, JWT, Google Cloud TTS |
| Scale | RESTful APIs managing 500+ government schemes |
| Performance | Sub-200ms response times via MongoDB indexing |
| Security | JWT-based authentication across services |
| Impact | 95%+ relevance scores on AI recommendations; multilingual accessibility via Google Cloud TTS |
| Repository | github.com/kpcode11/FarmerSahayak |
Engineered a microservice architecture connecting a Python Flask AI service (LangChain + Ollama RAG pipeline over ChromaDB) with a Node.js backend, structured for low-latency retrieval and accessible, multilingual delivery via text-to-speech.
π₯ CureOS β Hospital Information System
A hospital management prototype delivering secure, real-time clinical dashboards for clinicians and administrators.
| Stack | Next.js 15, React 19, TypeScript, PostgreSQL, Prisma, NextAuth.js, Socket.io, Tailwind CSS |
| Scale | Multi-role clinical system (clinicians, admins) with audit trails |
| Performance | Real-time metrics via Socket.io + Prisma queries |
| Security | Fine-grained RBAC model with full audit trails, NextAuth.js authentication |
| Impact | Reliable server-side PDF generation (jsPDF) for clinical reporting |
| Repository | github.com/kpcode11/CureOS |
Spearheaded the system architecture on Next.js 15 and TypeScript, layering a fine-grained RBAC model with audit logging on top of real-time Socket.io-driven analytics for live clinical metrics.
Software Development Engineering Intern β Zenup Health, Malad, Mumbai
Feb 2026 β May 2026
Constructed a full-stack CRM enabling pipeline tracking; migrated 30,000+ records from legacy platforms with 100% data integrity.
- Overhauled the React booking UI and built a debounced search engine, accelerating frontend load times by 35%
- Created a secure, post-booking in-app chat with AI-based moderation to flag suspicious conversations
- Designed an async WhatsApp message queueing system using Interakt and Supabase for reliable delivery
React CRM AI Moderation Supabase Interakt
IT Department Website Coordinator β Don Bosco Institute of Technology, Mumbai
Jul 2025 β Present
Managed official IT website operations and audits, ensuring 99.9% uptime and reducing content approval time by 40%.
- Boosted website performance via systematic optimization, achieving 25% faster load times
Web Operations Performance Optimization Audits
Python Instructor Intern β L&T and Creast Collaboration, Mumbai
Aug 2024 β Feb 2025
Instructed 50+ students in Python and OOP; generated progress reports that improved learning retention by 35%.
- Orchestrated 3 coding exhibitions and competitions, driving student participation up by 60%
Python Teaching OOP Mentorship
| Recognition | Details |
|---|---|
| π Best Gameplay Award | DBIT ACM Teknack 2025 β 1st place out of 60+ teams |
| π₯ Hackathon Ashwamedh β Top 6 | AI crowd-detection system built with YOLOv8 & OpenCV, ranked Top 6 of 40+ teams |
| π― Smart India Hackathon (SIH) 2024 | Advanced through college-level internal competition |
Udemy (Hitesh Choudhary)


