Computer Vision Engineer

• Oct 2022 - Present

Machine Learning Engineer

• Aug 2020 - Sep 2022

Data Scientist

• May 2019 - Jul 2020

• Implemented YOLO architecture to detect and extract handwritten information from document images • Created and trained a CNN model to classify record images according to their dates with an accuracy of 91% • Trained a DL regression model to improve the perspective of documents in the images • Performed OCR using Google Vision api to retrieve information from printed document images • Built a pipeline for the images to go through in production and conducted dry run of the models on ambari cluster • Created a feedback mechanism to retrain all the models with the data wrongly classified in production Show less

Machine Learning Intern

• Dec 2018 - Apr 2019

• Created TF.Estimator models to predict the semantic similarity between 2 text strings with an accuracy of 92% • Used LSTM, GRU and CNN architectures to create an ensemble model with the help of hyper-parameter tuning • Built a data ingestion and preprocessing pipeline for the models using TensorFlow Transform and packaged it • Implemented sentiment analysis on feedbacks to classify them among positive, neutral and negative reviews • Used TensorFlow-Hub Modules with fine-tuned dense layers to get an accuracy of 93% on validation set Show less

Intern

• May 2016 - Jul 2016

• Built an application to detect and track multiple faces from live video feed with an accuracy of 74.8% • Trained a haar cascade with 5,000 positive and 5,000 negative images to detect faces from a live video feed • Developed a model for tracing objects once detected using Mean Shift algorithm and OpenCV library in Python • Built an application to detect and track multiple faces from live video feed with an accuracy of 74.8% • Trained a haar cascade with 5,000 positive and 5,000 negative images to detect faces from a live video feed • Developed a model for tracing objects once detected using Mean Shift algorithm and OpenCV library in Python