Vikash Prasad

I am involved in all things related to NLP. A couple of products are below: 1. Fraud detection: Using customers attributes we try and detect fraudulent transactions and pinpoint areas causing them. The tech stack is composed of : Pyspark+Hive+Graphframes(pyspark) 2. Customer attribute extraction: Extracting customer profiles like age, education,qualification, income, etc from ASR(transcripts) which can be used to target specific customer profiles.

Responsible for maintaining full-cycle of models of core data based products(Deep Learning based text classification, Automated Newsfeeds) to be used by different clients using the tech stack of Python, Azure Ml, Azure Data Factory, Github, Pytorch, Pandas, Scikit-Learn

Responsible for implementing,maintaining and deployment of NER models Conversational AI(Gainsight Sally). Apart from this, I was also responsible for implementing a full-cycle churn predictive models of 5-6 different customers.

As an intern I learnt the major concepts of Data Science and about how data science is applied in Gainsight. Completed projects on: Contribute to building, testing and deploying machine learning models(September,2015-April,2016). Qualified for Galvanize BootCamp in Data Science(May,2016-August,2016). Project on Netflix Challenge Data(K-Means Clustering)

Mathematics, Statistics

Mathematics, Computer Science, Statistics