Training Model...

Epoch: 0/10
Accuracy: 0.00%
Loss: 2.3024

Hi I'm Aditya Dubey

Delivering AI Solutions

Passionate about building things that make a difference—whether it's through clever AI or solid software engineering. I love tackling complex problems and collaborating with others to create innovative solutions. Always learning and excited to contribute to impactful projects.

Connect With Me

Experience

My professional journey in machine learning

Feb 2025 - Present(June Expected)

STEP Intern

Maruti Suzuki India Limited

Optimizing real-time data ingestion pipelines Cloud Data Platform using AWS Glue and Kinesis Streams, with 22% improved scalability and throughput.

Jan 2024 - April 2024

Research Intern

IIT Patna (Indian Institute of technology)

Curated and preprocessed a large-scale dataset of over 3000 images for training and evaluation of the transformer model.
• Experimented with ViT architectures, adapting the Transformer model for image recognition tasks

Projects

CNN Project

Agentic Debate Simulator

Developed an AI-powered debate application using LangGraph for turn-based control and Flask with WebSockets for real-time backend-frontend communication for dynamic inference. Balanced GPT-3.5-turbo (cloud) and Mistral-7B-v0.1 (local) models.

Music Classifier

MorphoGAN

Image style transfer tool that transforms your content images into artwork inspired by famous artistic styles. Built using VGG19, Generative Adversarial Network and Adaptive Instance Normalization (AdaIN)

CNN Project

StoryScape

Django-based web application designed to help users generate creative stories, characters, and dialogues. It leverages advanced natural language processing (NLP) and AI-driven tools, such as GPT-2 and Google's Gemini API, to automate the process of storytelling.

CNN Project

Music Genre Classifier

MLP implementation of music genre using GTZAN Dataset. The model uses a dataset of audio features extracted from various music tracks, and it is trained to predict the genre of a given piece of music.

Publications

Jan 2025

Global Energy Consumption Patterns and Optimization using Big Data

2024 9th International Conference on Communication and Electronics Systems (ICCES)

Aditya Dubey; Anuj Tiwari; Ayush Bhardwaj;

Leveraging Apache Spark, Hadoop, and ML models (LSTM & Random Forest) to enhance energy consumption forecasting with 88% accuracy.

READ PAPER
Feb 2025

Conditional GANs: How to Control Generated Outputs

DataX Journal Medium Article

Aditya Dubey

This blog post will explore the architecture, mathematics, and applications of Conditional GANs in depth allowing for more controlled and targeted outputs of GAN process.

READ PAPER

Clubs

Organizations I contribute to

Data Science Community SRM

A technology driven Data Science-based student-led innovation community at SRM IST.

Visit Website

Next Tech Lab

The first and only student run multi-disciplinary lab of SRM University at Chennai and Amaravati.

Visit Website