Aniruddh Sharma
@aniruddhsharma
Data Scientist at Clientell Technologies
Bangalore, Karnataka
Aniruddh is a Data Scientist with advanced expertise in Machine Learning, Deep Learning, Natural Language Processing (NLP), and Computer Vision. He has significant experience in data gathering, cleaning, and organizing data. His background includes developing complex models, building robust ETL pipelines, and generating actionable insights using tools like AWS Quicksight.
Experience
Data Scientist
Clientell Technologies
Developed model for topic classification of e-mails through NLP techniques such as sBERT (sentence transformers), cosine similarity etc. and achieved 93% accuracy with 92% F-score on 42 topics. Developed model to extract contact information of all persons mentioned in e-mails through RegEx, NER via HuggingFace Transformers and StanfordCoreNLP models with accuracy of 97% over 250k+ emails. Built a model to detect if the e-mail is meeting related and the timings of the next/rescheduled meeting through Logistic Regression, BERT and synthetic sampling techniques such as ADASYN & SMOTE to achieve overall accuracy of 91% and f1-score 88%. Generated charts on dashboarding tool - AWS Quicksight acc. to client's needs based on their Salesforce/CRM dataset. Built ETL pipeline from data retrieval (AWS RDS) to automatic updating of dashboards at regular intervals on AWS.
SDE Intern (Frontend)
Techture Structures
Developed assigned parts of the product's website. Debugged and fixed the issues found in webpages through CSS, Javascript, & AngularJS.
Computer Vision Intern
Adra.ai
Assisted dental professionals with identifying issues with higher precision in dental radiographs to prescribe better treatment options. Built deep learning model for object detection on 25k+ dental radiographs.
SDE- Intern (Backend)
Reexis Systems
Improved Workforce Management System's performance and quality by debugging and fixing issues through Java, AngularJS, SQL.
Machine Learning Engineer
Repup
Improved Key-Phrase Extraction model, documented the source code and built NER model with 89% accuracy on online travel agency reviews. Tweaked the model's accuracy of sentiment classification from 92.6% to 94.8% through LSTMs and GRUs.
Machine Learning Intern
National University of Singapore (NUS)
Developed Real-Time Automatic License Plate Recognition (RT-ALPR) model. Achieved 86% accuracy with deep learning (YOLO), Computer Vision & OCR.
Education
Birla Institute of Technology And Science
B.E + M.Sc.