Mohit Patni

• Leading a GEN AI project as Middleware Lead .
• Implementation of chat summarization between customers and virtual agent. A chat summary is generated based on the conversation transcript between the customer and virtual agent .Using the embedded model GTP3.0
• Chat summarization for interactions between customers and human agents. Gen AI, generates concise summaries based on the chat transcripts between customers and human agents.Using the embedded model GTP3.0
• Next Best Response : When Customer is in conversation with an Agent . Gen AI generates the next best response on every message which customer types.
• Worked with customers or prospective customers to develop integrated solutions and lead detailed architectural dialogues to facilitate delivery of comprehensive solution.
• Guided and influenced existing partners on recommended upgrades and enhancements to integrated solutions.
• Created an end to end Cloud Agnostic framework to help existing insurance customer to start their migration journey to Cloud.
• Displayed sound understanding of standard networking protocols, virtual networks and load balancing.
• Provided architectural leadership and guidance to technical teams to deliver robust, highly scalable and cost-effective solutions.

►Developed a python spark based framework which will allow the end users to create the data pipelines just by creating the Metadata .

►Created a validation Framework to perform data quality checks and generate report for Count, Column to column check on sample data and Aggregate Checks(Min , Max and Null) .

►Created a Reporting Framework using ELK stack which gives users the ability to know the current status of Jobs running and validation reports in the form interactive visualizations.

►Worked as Tech lead for Spark Upgrade from 2.3.0 to 3.0

►Worked with apache spark team to increase the performance of data pipelines extraction from RDBMS. Able to load ,extract and validate the table with 2 billion records with 100 columns in under 2 hours.

►Worked with AWS team to setup exact services which we were using in EMR cluster on a EC2 Cluster. This saves lot of money for the firm since we used EC2 for our dev and UAT boxes.

►Created an CICD solution for deploying the code in dev , Uatand Prod by running in memory test case and black duck scan for the library scan with Jenkins.

Purpose of the Project: : Purpose of the project is to create a new analytics platform which can create the data pipelines on the go with user inputs from UI.

· Developed an end to end Big data analytic and reporting platform(OoBA) from scratch using Amazon Elastic search and custom Kibana.

· Developed a highly scalable ETL data process on AWS EMR to integrate data from multiple sources like Kafka, Elastic search, MongoDB, RDBMS(MySQL , oracle and PostgreSQL and Teradata ), S3, AVRO, JSON and external hive system .Also giving functionality on the UI to join these different sources on the go and finally load the data into data lake i.e. S3 or Elastic search.

· Implemented a Back end Data Quality module to improve and monitor the data quality of provided data sources using AWS Lambda which responds to events and automatically manages the underlying compute resources.

· Created a dev and UAT env by setting up the exact services which we were using in EMR cluster on a EC2 Cluster. This saves lot of money for the firm since we used EC2 for our dev and UAT boxes.

· Responsible for a smooth end to end development and deployment of data-driven products as highly robust REST services using Golang and Python(Pyspark ). These Rest Services were deployed on an EMR machine and gave the user’s ability to read the crunched data from data lake (s3 and Elastic search).

· Fulfilled all data engineer duties for new Analytics platform.

· Responsible for a smooth end to end deployment of data-driven products as highly robust REST services involving Golang as well as Python.