Seasoned Data Scientist & Machine Learning (ML) Engineer with experience in ML lifecycle, Data Engineering, and Cloud Technologies. Proficient in the development of production-level Machine Learning and Data Pipelines, mathematical research, and ML Research, particularly in Geometric Deep Learning.Hire Sanjeet
Developed Data Integration pipeline, e-commerce KPIs and dashboards, delivered production level ML pipeline for three algorithms: AR Background Remover, Doc2vec recommendation, and E-commerce Web Scraping as microservices through hands-on development and supervision of an ML team of 7 people, developed and deployed Neural Graph based recommendation models for e-commerce clients, and developed data collection servers.
Developed data flow architecture for NLU Principal Research and Innovation Project, analyzed and restructured data using feature engineering for efficient storage and usage, and set up back-end services of real-time database for a conversational Golf Caddie app of Hello Birdie on Google Cloud Platform, using Firebase, Dialogflow, and Cloud Functions.
Researched Machine Learning and Deep Learning techniques along with their connections to geometry. Published a paper on the stability of neural networks with delay feedbacks. Other key projects include: part-of-speech tagging through machine translation using NLP ecosystem, K-means clustering Spark pipeline, sentiment analysis of Twitter tweets using RNN GRU and CNN (softmax activation), Exploratory data analysis on suicide rates worldwide using Dataiku Data Science Studio, NLP Fasttext cbow and skipgram implementation of “Word embeddings” of Wikipedia dataset, Visualization and analysis of SNCF station data via Dash, Multiclass classification of Italian wine data from UCI website using Multinomial Logistic Regression, and Speech emotion detection using Multi Layer Perceptron and optimization by Stochastic Gradient and Adam