Akshat Bokdia | AI ML Enthusiast

Hello, I'm Akshat Bokdia.

A

Results-oriented AI/ML enthusiast with a solid foundation in statistics, programming, and machine learning, ready to dive into complex challenges and turn data into actionable insights

About

I am a final-year student at Vellore Institute of Technology (VIT), Vellore, pursuing a Bachelor's degree in Computer Science and Engineering. As a passionate AI/ML enthusiast, I am deeply interested in applying my analytical skills, programming proficiency, and problem-solving abilities to develop innovative solutions in the fields of Machine Learning and Data Science.

Proficient in Python, R, and SQL, I am committed to staying ahead of the curve in AI/ML trends and technologies and eager to contribute to cutting-edge projects that drive data-driven insights and transformative advancements.

Experience

Cheeni Labs

ML Project Intern

Fine-tuned a pre-trained General Embedding model for movie review classification
Pre-processed and optimized a 100,000-review dataset for training through a BERT-based model for a summary classifier for their product, Flixjini
Tools: Python, SQL, Sentence Transformers

Aug 2023 - Sep 2023 | Chennai, India

Prasath Lab, University of Cincinnati

Research Intern

Developed hybrid feature extraction techniques to enhance brain tumor classification in medical imaging
Built deep learning models to improve context-based retrieval accuracy for brain tumor images
Tools: Python, TensorFlow, OpenCV

Mar 2024 - Aug 2024 | Cincinnati, USA - Remote

Gerstein Lab, Yale University

Summer Research Intern

Trained CNN and Autoencoder models for structural variant classification in gene sequences
Preprocessed sequences and created embeddings for accurate classification
Tools: Python, TensorFlow, Hugging Face

Jul 2024 - Aug 2024 | New Haven, USA - Remote

Projects

Wave Control

A gesture recognition system that controls YouTube playback using hand gestures

Accomplishments

Tools: Python, OpenCV, TensorFlow
Perform actions like play/pause and volume adjustments
Interact with YouTube without physical contact

DebateMate

An AI-powered debate partner

Accomplishments

Tools: Python, Streamlit, GenAI
Enhance your debating skills with personalized experiences for all levels
Choose your topic, engage in dynamic discussions, and receive real-time feedback

Bully Detector

A system to accurately detect cyberbullying in text data

Accomplishments

Tools: Python, NLTK, Scikit-learn, MySQL
Deployed a machine learning model to detect cyberbullying in tweets
Helps mitigate harmful content on social media platforms

Uber Data Analysis

A project to analyze and visualize Uber ride data, uncovering trends and patterns

Accomplishments

Tools: R, RStudio
Conducted a comprehensive analysis of Uber ride data from April to September 2014
Developed visualizations using ggplot2 to uncover patterns and trends in ride behavior