Open to Opportunities

Software Developer & Engineer

I build modern applications — combining clean code, logical thinking, and practical problem solving.

Scroll
0+

Projects Done

0+

Years Exp.

Turning Ideas Into
Digital Reality

Hey! I'm Prem, a passionate Software Development Engineer (SDE) based in India. As a Computer Science student with strong foundations in Java, backend development, and full-stack technologies, I focus on building practical, scalable, and user-centric web applications.

I believe great software is built through strong problem-solving skills, clean architecture, and continuous learning. When I'm not coding, you'll find me practicing DSA, improving my system design knowledge, or building real-world projects that challenge me to grow as a developer.

Prem Kumar
2+ Years
Experience

Selected Work

Things I've shipped.

01
Full Stack
Smart Trip Planner
Problem Statement
Travelers waste hours stitching together flights, hotels, and activities from multiple platforms. There was no single tool that intelligently builds a day-by-day itinerary factoring in budget, travel time, and preferences.
Key Features
  • AI-generated personalized itineraries
  • Real-time flight & hotel cost aggregation
  • Collaborative trip sharing with live edits
  • Budget tracker with category breakdown
Architecture Summary
React Client Express API OpenAI
PostgreSQL + Redis Cache Vercel CDN
Tech Stack
React Node.js Express PostgreSQL OpenAI API Redis Docker
Challenges Solved
  • Rate-limited external API calls via Redis queue
  • Conflict-free collaborative editing with optimistic UI
  • Prompt engineering for structured itinerary JSON output
Impact / Learning
~60% reduction in planning time for test users
REST + WebSocket hybrid architecture for real-time sync
Scroll

End-to-end trip planner using AI to generate personalized itineraries with real-time pricing data and collaborative editing.

Full Stack AI Integration Real-time
02
AI / Computer Vision
VisionTrack AI
Problem Statement
Manual review of surveillance footage is expensive and error-prone. Security teams needed automated real-time detection of anomalous behavior and object tracking across multi-camera setups.
Key Features
  • Multi-object tracking across video frames
  • Anomaly detection with configurable thresholds
  • Live dashboard with alert notifications
  • Exportable incident reports with timestamps
Architecture Summary
Video Stream YOLOv8 DeepSORT
FastAPI WebSocket React UI
Tech Stack
Python YOLOv8 OpenCV FastAPI React WebSocket MongoDB
Challenges Solved
  • Real-time inference at 30 FPS on CPU with frame batching
  • ID re-assignment after occlusion using DeepSORT kalman filter
  • Async pipeline to prevent frame drop under load
Impact / Learning
94.2% tracking accuracy on test dataset
3× faster incident detection vs manual review
Scroll

Real-time video analytics system using YOLOv8 and DeepSORT for multi-object tracking and anomaly detection across camera feeds.

Computer Vision Python Real-time ML
03
Data / SQL
Retail Sales Analysis
Problem Statement
A retail chain with 5 years of transactional data had no structured way to identify revenue trends, seasonal patterns, or underperforming SKUs. Business decisions were made on gut feel.
Key Features
  • Cohort analysis for customer retention trends
  • YoY & MoM revenue comparison with window functions
  • Product category profitability heatmaps
  • Automated executive summary generation
Architecture Summary
Raw CSVs PostgreSQL dbt Models
Metabase + Python Reports
Tech Stack
PostgreSQL dbt Python Pandas Metabase Matplotlib
Challenges Solved
  • Handled 12M+ rows with indexed CTEs for sub-second queries
  • Normalized inconsistent date formats across 5 source systems
  • Idempotent dbt transformations with incremental materialization
Impact / Learning
12M+ rows processed, queries under 800ms
SQL Window Fns deep expertise in analytical SQL patterns
Scroll

End-to-end SQL analytics pipeline over 12M+ retail transactions, surfacing cohort trends, SKU performance, and executive dashboards via dbt + Metabase.

SQL dbt Data Analytics
04
Full Stack
Online Car Rental System
Problem Statement
Local rental agencies operated on paper-based booking systems, leading to double-bookings, lost revenue, and poor customer experience. Needed a fully digital, self-service rental platform.
Key Features
  • Date-aware availability engine with conflict prevention
  • Role-based admin panel for fleet management
  • Integrated payment with invoice generation
  • Email notifications for booking lifecycle events
Architecture Summary
React SPA Django REST PostgreSQL
Stripe + Celery / Redis AWS S3
Tech Stack
React Django PostgreSQL Celery Redis Stripe API AWS S3
Challenges Solved
  • Atomic DB transactions to prevent race-condition double-bookings
  • Async email queue via Celery so checkout stays fast
  • Dynamic pricing engine based on season and vehicle class
Impact / Learning
Zero double-bookings in 6 months of testing
Atomic txns deep dive into DB concurrency & locking
Scroll

Full-stack car rental platform with atomic booking logic, Stripe payments, async email notifications, and an admin fleet management panel.

Full Stack Django Payments

The Journey

Three years of becoming.

Hover over a year to explore →

  • Joined the Design Club

    Dove into UI/UX fundamentals, contributed to campus event branding, and discovered a love for intentional aesthetics.

  • First Hackathon — Top 10 Finish

    Shipped a working prototype in 24 hours; learned rapid iteration and how to communicate an idea under pressure.

  • Mastered HTML, CSS & JavaScript

    Moved from tutorials to building real things — personal projects, UI experiments, and countless redesigns.

  • Built My First Portfolio

    A rough v1, but it was mine. Learned version control, deployment, and the value of showing your work.

Skills & Technologies

Frontend

HTMLCSSJavaScriptResponsive DesignDOM ManipulationUI Design
🔧

Backend

JavaNode.jsExpress.jsREST APIsAuthenticationServer-side Logic
☁️

Database

SQLMySQLMongoDBDatabase DesignQuery OptimizationData Modeling
🎯

Core CS & Problem Solving

Data StructuresAlgorithmsOOPDBMSOSComputer Networks
Java (DSA & OOP)92%
Backend Development (Node.js / Express)70%
Frontend Development (HTML / CSS / JS)88%
Database Management (SQL / MongoDB)83%
Problem Solving & DSA70%
Java Python SQL HTML CSS JavaScript MongoDB Express.js Node.js Git Java Python SQL HTML CSS JavaScript MongoDB Express.js Node.js Git

Let's Build
Something Great

Have a project in mind? I'd love to hear about it. Send me a message and let's create something amazing together.

📍 India , AP
🕐 Available Mon–Fri, 9am–6pm IST