AI Engineer • Data Scientist • ML Researcher

Building Intelligent
Digital Experiences

Hi, I’m Deepayan Thakur - a Machine Learning Engineer & Data Scientist passionate about creating AI systems, forecasting platforms, NLP pipelines, and futuristic intelligent products that solve real-world problems.

Person with VR Headset m_f
Minimalist Stone Composition m_f
Whimsical Character in a Jar m_f
Three Scoops of Ice Cream on Spoons
Green Character in Yellow Hoodie m_f
Explore Projects
Let’s Connect

Intelligence Architecture

Specialized domains in Artificial Intelligence and Machine Learning.

Deep Learning

Architecting Convolutional Neural Networks (CNNs) for complex image classification and pattern recognition tasks.

ConvNeXt ResNet

Predictive Analytics

Building time-series forecasting engines for large-scale economic data, utilizing statistical and neural models.

LSTM ARIMA

Computer Vision

Implementing real-time data augmentation and binary classification for medical imaging and cryptography.

Data Augmentation OpenCV

Latest Masterpieces

A collection of AI models, web apps, and data analytics platforms.

Medicinal Plant App
Web & CNN

Medicinal Plant Detection

Developed a Web Based app utilizing ConvNeXt Tiny CNN to classify 40+ medicinal plants with real-time camera input and health benefits UI.
Cavity Detection
Imaging

Teeth Cavity Detection

Built a CNN model to detect cavities from dental X-ray images. Used data augmentation and binary classification for early detection.
DRDO Security
Crypto / DL

DRDO Image Classify

Combined cryptographic techniques with CNN-based classification for secure image transmission. Ensured data integrity in defence datasets.
DPIIT Analytics
Forecasting

Trade Analytics & Forecast

DPIIT project. Automated trade data fetcher, multi-model forecasting engine (ARIMA, Prophet) achieving 93% accuracy for FY25.

Metrics & Logs

Coding statistics and recent technical writings.

1,250+

Contributions

200+

Problems Solved

93%

Model Accuracy

40+

Plant Classes

Technical Log

Architecting a ConvNeXt Tiny Classifier

A deep dive into optimizing lightweight CNNs for real-time mobile inference.

 Case Study

Time Series Forecasting with Prophet & LSTM

Comparing statistical approaches vs deep learning on DPIIT trade data analytics.

About Me

Pursuing a B.Tech in CSE (AI & ML) as a specialization from Sharda University. My journey is defined by a passion for scalable AI solutions and solving real-world problems.

With hands-on experience from DPIIT, DRDO, and NHAI internships, I am skilled in Python, Deep Learning, and Data Science. I am dedicated and eager to contribute to an innovative team.

Python SQL Java & C HTML & CSS Deep Learning Data Analytics

Academics

  • B.Tech CSE-AIML

    Sharda University

    8.808

    2026

  • Intermediate (XII)

    KV, New Delhi

    83.8%

    2022

Verifications

AI & Machine Learning Data Analyst - Space Tech Python Programmer

Professional Journey

Data Analyst Intern

Jun '25 - Aug '25

DPIIT, New Delhi

Developed an automated HSN-based Trade Data Fetcher, chatbot, and dashboards. Implemented forecasting models (LSTM) achieving 93% accuracy for FY25 toy sector predictions.

Machine Learning Trainee

Jun '24 - Jul '24

DRDO, New Delhi

Focused on the development of a deep learning model for classifying encrypted data based on cipher algorithms, including TDES and AES, as part of 'Cipher Algorithm Classification Using Deep Learning'.

Data Integration Engineer

Jun '23 - Aug '23

NHAI, New Delhi

Developed a 'Data Integration and Management System' using Python and SQL Server. Integrated CNN for pattern recognition, streamlining large-scale infrastructure data analytics.