Saumya Thakor - AI/ML Engineer Transforming Intelligence
into Impact

Solving real-world problems using machine learning, Deep Learning, and computer vision

Welcome! I'm a developer of intelligent systems. I design models, scale architectures, and build real-time apps. Let's explore my work and create something incredible together.

8+ Projects
10+ Skills
1 Publication
Saumya Thakor
Scroll Down

About About Me

Education

Current

M.Tech @ IIITM Gwalior

2024–2026 | CGPA: 8.74

Specializing in AI/ML and Data Science

Completed

B.Tech CSE @ PDEU Gandhinagar

2020–2024 | CGPA: 9.39

Computer Science & Engineering

Who I Am

A highly motivated AI/ML Engineer with a passion for transforming real-world problems into scalable solutions using AI, ML, and deep learning. Strong in building models, developing intelligent systems, and working across domains.

Machine Learning
Deep Learning
NLP
Computer Vision
AI Enthusiast
Wireless Networks
Data Analysis
Image Processing

Technical Skills

AI/ML Technologies

Deep Learning
NLP
OpenCV
Scikit-learn
LangChain
PyTorch
TensorFlow

Tools

Docker
Git
Flask
Streamlit
Render
Tableau
Power BI

Programming Languages

Python
SQL
PostgreSQL

Projects Featured Projects

AI Travel Itinerary Generator

An intelligent travel planning application powered by Groq 80B model via API. Takes destination preferences and generates detailed, personalized itineraries with LangChain integration.

Takes place + preferences for detailed itinerary
Powered by Groq 80B model via API
LangChain + frontend integration

Oblivion-Qwen Chatbot

A powerful chatbot built by locally fine-tuning the Qwen base model using QLoRA and the Alpaca dataset. Optimized on personal hardware, it delivers intelligent, context-aware conversations with a sleek UI, offering strong Q&A capability.

Custom fine-tuned QLoRA model
Efficient Token Handling
Interactive Chatbot UI

NSE Stock Predictor

Advanced stock prediction system using hybrid LSTM-CNN architecture. Trained on comprehensive dataset of 2126 Indian stocks with sophisticated UI and multiple analytical features.

LSTM-CNN hybrid model for stock prediction
Trained on 2126 Indian stocks
Advanced UI with multiple features

ANPR with OpenCV

An intelligent computer vision system for accurate license plate detection and recognition using image processing and EasyOCR. Efficiently reads plate numbers from real-world scenes in real time, ideal for surveillance, traffic, and security.

Automatic Number Plate Recognition system
Advanced Image Processing Pipeline
Robust Text Extraction with EasyOCR
Data Science Visualization

Improving Spectral Clustering Scalability through Intelligent Sampling Methods

Research project on improving spectral clustering scalability through intelligent sampling methods. Published in EAI International Conference on Body Area Networks.

Intelligent sampling algorithms
Scalability improvements for large datasets
Published research findings

ML-Based Task Load Balancing for IoT Devices

A smart task allocation system that uses machine learning to dynamically balance computational loads between IoT devices and cloud servers. It evaluates task complexity, device status, and energy constraints to make real-time decisions, improving efficiency and reducing latency.

Intelligent task offloading between edge and cloud
Supports multiple IoT platforms (Raspberry Pi, ESP32, Arduino)
Advanced performance visualization and analytics

Fake News Detection

A high-performance NLP system for real-time fake news detection, using advanced machine learning and text analysis. Delivered via a robust Flask web app, it ensures good accuracy and user-friendly misinformation detection.

NLP-based fake news classifier
Advanced NLP Pipeline
Instant Classification
Medical AI

Cancer Image Classification

Advanced medical image classification system using ensemble learning. Combines DenseNet, VGG16, and InceptionResNet architectures with sophisticated data augmentation techniques.

DenseNet, VGG16, InceptionResNet ensemble
Advanced data augmentation techniques
High accuracy medical image analysis

Research Publications

Improving Spectral Clustering Scalability through Intelligent Sampling Methods

EAI International Conference on Body Area Networks (EAI BodyNets) Accepted Publication: To be published by Springer as part of the conference proceedings.

Conventional spectral clustering methods provide essential information about the structure within the dataset; nonetheless, they are not scalable over large datasets. In this research, an ensemble of density-based and cluster-based sampling techniques is used to improve the scalability of spectral clustering in a novel way. By carefully choosing a representative sample of data points, the suggested strategy lowers computing complexity while speeding up the clustering process without sacrificing accuracy. Our tests show that the suggested approach performs better in terms of clustering performance (as determined by the Silhouette Score, Adjusted Rand Index, and Normalized Mutual Information) and computing efficiency than conventional spectral clustering and other cutting-edge approaches. In large-scale datasets, the approach reduces execution time by 134.4\% while improving clustering accuracy by 61.5\%. This method is potentially used in large-scale data analysis applications where scalability and efficiency are crucial, including body area networks (BANs).

Timeline My Journey

2020

Started B.Tech @ PDEU

Began my journey in Computer Science & Engineering

2023

Summer Internship & Minor Project

ML Intern @ Inrainz (Remote) – Fraud Payment Detection
Developed first major Deep Learning project

2024

GATE, Major Project & Internship

Completed B.Tech and cleared GATE
Major project in final semester
IT Support Intern @ ChemProcessSystem (Ahmedabad)

2024–26

M.Tech @ IIITM Gwalior

Currently pursuing specialization in AI/ML

Dec 2024

Research Presentation @ IIT BHU

Presented at EAI BodyNets 2024, IIT BHU – Springer publication

2025

Advanced AI Projects

Did several AI projects and research work

Connect Let's Connect

Open to Opportunities

I’m open to full-time, internship, and collaborative roles in AI/ML, data science, and tech innovation. Let’s connect to explore how I can bring value to your team.