Well, well, well, looks like someone's curious about me!
Well, I was born in 1999 in India and have always had a deep interest in robotics and artificial intelligence.
After completing my B.Tech in Electronics and Communication Engineering in 2020, where I explored embedded
systems like Arduino and Raspberry Pi, I shifted my focus to machine learning and data science.
I embarked on a journey of self-learning and hands-on projects, eager to delve deeper into this field.
In 2021, I joined Avaamo.ai as a Conversational AI Intern and quickly progressed to become a Conversational AI Engineer within a year.
During my time there, I developed intelligent chat agents integrated into websites, leading to a significant 30% improvement in user navigation efficiency.
Determined to further expand my knowledge, I pursued a Master's degree in Data Science in the USA and graduated
from the New York Institute of Technology in 2023. During my master's program, I immersed myself in statistical
machine learning models in the first year and explored advanced technologies in the second year.
These included Deep Neural Networks, Image Classification, Segmentation, Object Detection, Recurrent Networks,
Language Models, Transformers, and Autoencoders.
My journey in artificial intelligence reflects a deep passion and commitment to mastering the field.
I believe in the power of connections and the value of sharing knowledge. If you share my enthusiasm
for AI or simply want to connect, please don't hesitate to reach out. Let's start a conversation, I'm curious about you too!
Urban Street Scene Understanding
Employed PyTorch for UNet and SegNet models on the Cityscapes data set, achieving 90% accuracy in semantic segmentation. Highlighted deep learning and DCNN expertise in urban scene analysis.
BBC News Articles Clustering
Leveraged K-means and advanced NLP for clustering 2225 BBC News articles, applied PCA for dimensionality reduction and LDA for topic naming, and visualized interactive results with t-SNE and Plotly.
Lung X-ray Segmentation
Developed a UNet Encoder-Decoder for lung segmentation in X-ray images, applying advanced processing methods to enhance quality and achieving an 82% IoU score, demonstrating proficiency in image segmentation.
CIFAR-10 Image Classification
Engineered a DCNN with Residual and Inception blocks for CIFAR-10, reaching 90% accuracy. Optimized with callbacks, and learning rate schedulers, and integrated into a Telegram bot.
ML-based Diagnosis Website
Designed a Django-based medical diagnosis web platform, integrating traditional ML models using scikit-learn and SQLite database for data handling, deployed on AWS EC2 for enhanced scalability and reliability.
Cow Teat Mastitis Classification
Built a custom ResNet model using PyTorch for cow teat mastitis classification, categorizing severity into 4 levels. Implemented pseudo-labeling for data augmentation, achieving 85% accuracy.