About

Govind Daga is an experienced Machine Learning Engineer with a strong background in Data Analysis, Machine Learning, Deep Learning, and Databases. He has a Bachelor's degree in Computer Science from Jawaharlal Nehru Technological University, Kakinada. Currently, Govind is working as a Software Developer at Amdocs where he has automated flows of previously manual billing routines for Mobile and Internet services for AT&T using shell and python. He has also created new flows to estimate revenue and cron jobs to track critical files and data that are essential for One Click Billing. Additionally, he has developed SPOG dashboards to monitor KPIs and health checks of products and their applications using Grafana and InfluxDB. He has also exported this data to different personnel to run ML models for Anomaly Detection and Clustering and alert and prepare for various circumstances. In his previous role as a Machine Learning Engineer at PurpleTalk, Govind developed a Q&A Generation NLP Bot by fine-tuning Transformers (BERT) to generate Questions and Answers based on internal documents about their products and services for Serviceaide, Inc. He also performed Keyword Extraction using RAKE and SpaCy on the data from the documents. Initially, he trained and tested a model by performing NER, POS Tagging, and Word2Vec with custom RNN. For Dr Lal Path Labs, he performed EDA by recognizing correlations and covariables and plotted and observed trends on test time patterns, and performed other feature engineering tasks. After EDA, a custom Neural Network was trained to achieve 72% Accuracy and this model resulted in a 30% decline in IVR calls for the Client. He also performed Affinity Analysis on the tests taken by the patients and created a Recommendation Model to recommend Frequently-Bought-Together tests. Prior to this, Govind worked as a Data Analyst at PurpleTalk where he worked on a project to optimize the workflow of Sample Testing Logistics of Dr Lal Path Labs using Deep Learning and Pattern Recognition. He collected SQL Dumps from multiple tables and organized them into 8,000,000+ samples belonging to various tests like Liver & Kidney Panel Test, HIV 1 & 2 Antibodies tests, etc. from thousands of patients’ records. He recognized patterns of data irregularities and bad data and automated the data cleaning using PySpark. He also applied Transformations to the data from the tests and specifications of the labs to pre EDA data format. Gov