Priyanshu Shukla

PRIYANSHU
SHUKLA

I build things that ship to real users.

View Resume
450+
DSA Problems
3
Live Apps
1
Patent Filed
300+
Students Automated
⚛ React
🐍 Python
🎸 Django
▲ Next.js
🍃 MongoDB
👁 OpenCV

B.Tech CSE @ IIMT College of Engineering (2022–26). I turn ideas into production-grade software — from AI-powered attendance systems to founder-led e-commerce platforms. I don't just code; I ship.

Open to opportunities
0+

DSA Problems Solved

0

Live Production Apps

0

Patent Filed

0+

Students Automated Daily

Builder First

Every project I take on has real users. From a 500-face/day attendance system to an e-commerce platform, I prioritize shipping over perfection.

Research & Innovation

Patent holder for LUMIFACE — an AI-powered facial recognition attendance system. Filed under App No: 202511051742 A.

Competitive Edge

ICPC Asia Kanpur Regionals participant. 450+ DSA problems solved across LeetCode, GFG, and CodeChef.

Experience: Bharat Intern

Dec 2023 – Jan 2024. Built robust frontend architectures focusing on responsive web development, robust REST API integrations, and agile Git collaboration.

Python·Django·Next.js·React·Three.js·TypeScript·JavaScript·Node.js·OpenCV·MongoDB·MySQL·PostgreSQL·JWT·REST APIs·Machine Learning·Computer Vision·GSAP·Framer Motion·Tailwind CSS·HTML5·CSS3·Git·Docker·Vercel·AWS·Redis·GraphQL·FastAPI·TensorFlow·Python·Django·Next.js·React·Three.js·TypeScript·JavaScript·Node.js·OpenCV·MongoDB·MySQL·PostgreSQL·JWT·REST APIs·Machine Learning·Computer Vision·GSAP·Framer Motion·Tailwind CSS·HTML5·CSS3·Git·Docker·Vercel·AWS·Redis·GraphQL·FastAPI·TensorFlow·
Patent Filed — App No: 202511051742 A

LUMIFACE

Patent-pending AI attendance via facial recognition

Computer vision attendance system processing 500+ faces/day. LBPH + Haar Cascade pipeline achieves 75%+ accuracy. Reduced manual marking time from 15 minutes to 2 minutes.

500+Faces/Day
75%+Accuracy
15m → 2mTime Saved
PythonOpenCVMySQLLBPHHaar Cascade
Private Deployment GitHub
LUMIFACE_v2.0 · STATUS: TRACKING

AI Interviewer

Real-time AI-powered technical interview simulator

Conducts live technical interviews using OpenAI GPT-4. Generates adaptive questions, evaluates answers in real-time, and produces detailed feedback reports. Deployed for 100+ users.

GPT-4Adaptive Questions
< 1sReal-time Eval
PythonDjangoNext.jsOpenAI APIMongoDB
ai-interviewer-chi-henna.vercel.app/
Loading Live Preview
Open Live Site
⚡ Founder

Kazuha Closet

Live e-commerce platform — Founder & Full-Stack Dev

End-to-end fashion e-commerce platform live in production. Built the entire stack solo — from Django REST backend to Next.js storefront. 30% engagement uplift, 40% faster API response.

+30%Engagement Uplift
+40%API Speedup
Next.jsNode.jsDjango RESTMongoDBTailwind
kazuhacloset.vercel.app/
Loading Live Preview
Open Live Site

TaskFlow Manager

Drag-and-drop Kanban for engineering teams

Full-stack project management with real-time drag-and-drop Kanban boards, JWT auth, team workspaces, and deadline tracking. Improved workflow efficiency by 40%.

+40%Efficiency Boost
WebSocketsReal-time Sync
Next.jsDjangoMongoDBJWTTailwind
taskflow-manager-fullstack.vercel.app/login
Loading Live Preview
Open Live Site

Let's build
something.

Open to full-time roles, freelance projects, and exciting collaborations. Let's talk.

Send a Message →

Built by Priyanshu Shukla · 2025 · Next.js + Three.js