Kunj Pandya

Full-Stack Engineer & Entrepreneur

📍 Hoboken, NJ

Building scalable backend systems and intelligent applications. From distributed task orchestrators to AI-powered platforms, I turn ideas into production-ready solutions that scale.

10K+
Tasks/Day Processed
500+
Active Users
94%
ML Accuracy
View Resume

About Me

Full-stack engineer and aspiring entrepreneur passionate about building scalable systems that solve real-world problems.

Currently pursuing MS in Computer Science at Stevens Institute of Technology while architecting distributed backend systems, microservices, and cloud-native applications.

From building distributed task orchestrators handling 10K+ tasks/day to creating AI-powered platforms serving 500+ users, I focus on turning ideas into production-ready solutions that scale.

Education

Master of Science

Computer Science

Stevens Institute of Technology

Hoboken, NJ

Sept 2025 – May 2027

Key Coursework:

Distributed SystemsCloud ComputingConcurrent Programming+2 more

Bachelor of Technology

Artificial Intelligence and Machine Learning

D. J. Sanghvi College of Engineering, Mumbai University

Mumbai, India

Sept 2021 – May 2025

Key Coursework:

Machine LearningDeep LearningComputer Vision+2 more

Experience

Backend Engineer – Distributed ML Platform

Current
Stevens Institute of Technology
Hoboken, NJ

Nov 2025 – Present

  • ▸Architecting and building a distributed profiling platform for federated machine learning, handling energy metrics collection from 100+ edge devices in real-time.
  • ▸Developed scalable backend infrastructure using TensorFlow.js, Socket.IO, and Redis to process per-layer power consumption data across heterogeneous device networks, optimizing for low-latency data aggregation and fault tolerance.
TensorFlow.jsSocket.IORedisDistributed SystemsReal-time ProcessingEdge ComputingSystem Architecture

Founding Engineer – Placement Intelligence Platform

D. J. Sanghvi College of Engineering
Mumbai, India

May 2024 – June 2025

  • ▸Built and scaled backend infrastructure from scratch for a placement platform serving 500+ students. Designed RESTful APIs handling 10K+ requests/day with <200ms p95 latency, optimizing database queries and implementing efficient caching strategies.
  • ▸Architected and deployed ML inference microservices including an spaCy-based resume parser (95% accuracy) and real-time computer vision pipeline processing at 30 FPS. Managed end-to-end product development from initial prototype to production deployment.
Node.jsSQLiteRESTful APIsMicroservicesspaCyTensorFlowSystem DesignProduct Development

Featured Projects

A selection of projects demonstrating expertise in distributed systems, AI/ML, and scalable backend architecture

Full-Stack & Machine Learning

Dummy Trading — Paper Trading Platform

Feb 2026

Comprehensive paper trading platform with $100K virtual funds, real-time Alpaca pricing, ML-powered predictions, and competitive group trading.

Impact: Full S&P 500 market browsing with real-time pricing. ML prediction system with 23-feature engineering pipeline from OHLCV data. Supports limit/stop orders with automated 60-second validation cycles.

ReactTypeScriptNode.jsPostgreSQLPrisma+3 more
View on GitHub
View More Projects on GitHub →

Recent Posts

Thoughts on distributed systems, AI/ML, and building things that scale

March 2026

Just shipped a distributed task orchestrator that processes 10K+ tasks/day using Go + Redis Streams. The hardest part wasn't the throughput — it was designing the failure recovery logic so no task is ever lost, even mid-crash.

#Distributed Systems#Go#Backend
428
February 2026

Built a computer vision pipeline for real-time hand gesture recognition using MediaPipe + a custom LSTM model. Achieving 94% accuracy at 30fps on edge hardware. The secret: data augmentation + temporal smoothing.

#Computer Vision#ML#Edge AI
6712
January 2026

Reflected on building my first SaaS — the technical decisions that scaled and the ones that didn't. Microservices felt like overkill at user #10 but saved us at user #500. Context matters more than best practices.

#Entrepreneurship#SaaS#Architecture
8923
View full profile on LinkedIn →

Technical Skills

Languages

Java (Proficient; Concurrency)Python (FastAPI, NumPy)C/C++ (CUDA, Systems Programming)JavaScript/TypeScriptGo

Backend Systems

Distributed Systems (DHTs, Fault Tolerance)gRPCRESTful APIsPostgreSQL (Schema Design)RedisMicroservices

Infrastructure & ML Ops

DockerAWS (EC2/S3)Model Inference PipelinesLinux/BashCI/CD PipelinesGit

AI/ML

TensorFlowPyTorchComputer Vision (YOLO, ResNet)NLP (spaCy, NER)Deep LearningRecommendation Systems

Developer Tools

GitUnit Testing (JUnit, PyTest)Maven/GradleVS CodeDebugging & Profiling

Always learning and exploring new technologies to build better systems

Get in Touch

I'm always open to discussing new opportunities, collaborations, or just having a chat about distributed systems and AI/ML.

Email

kunjspandya@gmail.com

GitHub

@kpandya1903

LinkedIn

Connect with me

Location

Hoboken, NJ

© 2026 Kunj Pandya. Built with Next.js, TypeScript, and Tailwind CSS.