I love to work with data and visualise it to get useful insights and build fascinating stories. I tend to specialize in data analysis which boosts exposure and authority. Provided the right vision I can do wonders with data and impress you with my creative storytelling !
ResumeB.Tech. Biotechnology CGPA - 8.03
Amity Institute of Biotechnology, Amity University Noida
Class X - CGPA 10
Class XII - Medical with Mathematics - 88.2 %
Oxford Senior Secondary School, West Delhi
Worked on new text detection algorithm for curved text detection using computer vision and deep learning in recaptured and social media images used state of the art text detection models like EAST, TextBoxes++, PSENet etc.
Worked on GANs for generating replica of adult movies and then classify those videos as fake or real using transfer learning models like VGG16, ResnNet100 etc.
Worked on different projects including building recommendation systems, sentiment analysis based on reviews using NLP, chat bot using Seq2Seq architecture with attention mechanism and analysis of real world RBI Data.
Worked on text summarization wherein I applied different text summarization models like T5 Transformers, DistilBERT, AlBERT, XLNet etc. Build a NLP process pipeline and a pypi package for news article summarization.
Working under Dr. Kamal Rawal on his multiple DBT and US funded projects including – In-silico vaccine design for chagas disease, COVID-19 vaccine design by Reverse Vaccinology approach using Deep Learning and others.
Working as a data analyst for CogniTensor, working primarily on maintaining the dashboards for clients and writing generalised python code for topics like correlation for time series data and logging automation.
Working as a data analyst for CogniTensor, working primarily on maintaining the dashboards for clients and writing generalised python code for topics like correlation for time series data and logging automation.
VaxiDL is a deep learning based web server which is built in order to predict Whether a protein sequence (FASTA) or nucleotide is a potential vaccine candidate or not based on its biological features derived from Reverse Vaccinology Pipeline. It is hosted on web.
Build a multiclass classification deep learning model and trained on images of different anatomical and for Endoscopic regions of human Gastrointestinal tract e.g. pylorus, Z-line etc. Trained and tested all different CNN optimizers on images and used techniques like Data Augmentation & Transfer Learning (VGG16,InceptionNetV3 and EfficientNet) for training the model. For hyper parameter tuning used keras-tuner and got validation accuracy of around 94%.Deployed the model on pivotal cloud using Flask backend.
Build a deep learning model to predict and segment brain tumour in patient’s MRI scan. Used U-Net architecture for performing segmentation and VGG16, EfficientNet models for tumor prediction. Deployed the prediction model on pivotal cloud using Flask backend.
Built an integrated web application comprising of different ML models to predict health of a person it used models for kidney, heart, diabetes, cancer, liver, lungs (COVID) etc. disease detection all under a single web application.
Generative adversarial network models to generate fake images that looked real using different datasets including dataset of celeb faces and anime character faces.
This project is built using OpenCV library.It works by taking live feed from webcam and marks Attenance based on database of images of students. it records name and time of joining the class and saves it into csv file.
Build an employee attrition prediction model as a part of final project for Summer Analytics 2020 (IIT Guwahati). Built dashboard for EDA using Tableau. Built a ML classification model based on data and deployed the model on Heroku cloud platform using Flask backend.
Collected air quality data from different websites using Beautiful Soup Library, build different machine learning Models to predict PM 2.5 levels and deployed the final Model on Heroku cloud using Flask.
Collected live data from Reddit using PRAW and stored it into MongoDB. Built different machine learning classification models to predict the flair for the post along with that built a live tracking system to track the new post as spam or non-spam. Deployed the model on Heroku Cloud..
To Study the Effect of Unconventional Treatment Protocol on COVID-19 Patients in Delhi Using Artificial Intelligence Based Methods. (SSRN 18 September 2020)