Skip to content
OJASWI GUPTA

Architecting real-time systems and scalable intelligence.

Systems & ML Engineer specializing in Edge AI, distributed architectures, and high-performance applications.

Started with C++ and a curiosity for how things work under the hood. Fell in love with distributed systems and real-time architectures. Now building intelligence at the edge — fast, lean, and scalable. Vice President & Founding Partner of the Entrepreneurship Club. GSSoC'25 Tech Contributor · NEC Finalist · Builder at heart.
NLP & Machine Learning

GistProbe

An advanced NLP web analyzer that dynamically crawls, deduplicates, and clusters content using K-Means and FAISS RAG for Llama-3 abstractive summaries.

Python scikit-learn FAISS Flask Llama-3
View Details ↗
Llama-3
MODE: RAW_DATA
DOCUMENTS: 124
NLP Pipeline Engine
Edge AI & Hardware

Smart Waste Segregation Bin

Power-efficient embedded deployment on Raspberry Pi 4 using MobileNetV2. Optimized via ONNX runtime for motion-triggered inference (organic/plastic/metal/glass) using Pi Camera and LIDAR.

Raspberry Pi MobileNetV2 ONNX Python LIDAR
View Details ↗
CLASS: PLASTIC_BOTTLE
CONF: 98.4%
ONNX Inference Bin
Real-Time Systems

Live Video Chat Platform

Real-time Omegle-style web application utilizing WebRTC with STUN/TURN and WebSockets signaling. Features queue management and NAT traversal logic for seamless peer-to-peer connections.

WebRTC WebSockets Node.js Socket.IO React
View Details ↗
P2P CONNECTED
RTT: 28ms
WebRTC Signaling HUD
Full Stack ML

Pest Identification System

End-to-end classification pipeline featuring a TensorFlow model served via FastAPI. Delivers real-time 9-class pest identification through a REST API and a responsive web frontend.

TensorFlow FastAPI Python REST API JavaScript
View Details ↗
PEST_A 94%
PEST_B 4%
PEST_C 2%
FastAPI REST
FastAPI Classifier
Quantum ML

O-Finance

Explored quantum machine learning by implementing a Variational Quantum Classifier (Qiskit). Engineered financial indicators to predict FTSE 100 direction with competitive test accuracy.

Qiskit Quantum ML Python scikit-learn
View Details ↗
|ψ⟩ Qiskit
Qiskit Qubit Forecast

About

Ojaswi Gupta

Hi, I'm Ojaswi. I build highly performant, scalable systems that bridge the gap between heavy technical capability and seamless user experience.

With an academic foundation in Computer Science from VIT Amaravati, my focus lies in Machine Learning, Distributed Systems, and Real-Time Architectures. I specialize in taking algorithms from concept to production—whether that means optimizing neural networks for edge devices or designing precise real-time data processors handling massive concurrent loads.

Beyond code, I drive technical community growth as the Vice President & Founding Partner of the Entrepreneurship Club, a GSSoC'25 Tech Contributor, and an NEC Finalist.

Languages & Backend

Building the core engines with robust, typed, and scalable languages.

Python C++ Java TypeScript JavaScript SQL Node.js FastAPI Flask

ML & Infra

Deploying models to the edge with precision.

PyTorch TensorFlow scikit-learn ONNX Docker Vercel

Systems & Frontend

Real-time communication and seamless user interfaces.

WebSockets WebRTC Socket.IO REST APIs React Tailwind CSS

Visual Core Settings

Customize the portfolio's active accent color dynamically.

Journey

2023

Academic Launch & Community Leadership

Co-founded and drove leadership as Vice President and Founding Partner of the Entrepreneurship Club at VIT Amaravati. Focused on C++, high-performance structures, and distributed backend networks.

2024

National Recognition & Innovation

Staged as a National Finalist in the National Entrepreneurship Challenge (NEC). Led structural development and incubated products bridging systems engineering with financial prediction models (O-Finance).

2025

Open Source & Systems Growth

Contributed as a GSSoC'25 Tech Contributor. Partnered in designing decentralized models and WebRTC live media synchronization, while pushing architectural benchmarks to new heights.

2026

Building Intelligence at the Edge

Deepening specialization in compiling high-efficiency deep learning models for restricted edge computing environments and ultra-low-latency real-time signaling systems.

Let’s Build Something Meaningful

I’m always open to discussing new opportunities, collaborations, or ambitious ideas in software, distributed systems, and AI.

If you have a challenging problem worth solving, let’s talk.

← Back to Portfolio