Domain: Consumer finance company | Tech Stack: Python, Jupyter Notebook | May 2019
 Objective: Consumers to understand how consumer attributes and loan attributes influence the tendency of default
 Solution: Designed a machine learning model via EDA to understand how consumer attributes and loan attributes
influence the tendency of default. This analysis will help consumer finance company to decide if loan can be given to
applicants depending upon their credit history
 Key Achievement: Developed a model with an Insights of the consumer
Domain: Automobile company | Tech Stack: Python, Jupyter Notebook | May 2019
 Objective: A Chinese automobile company Geely Auto aspires to enter the US market by setting up their manufacturing
unit there and producing cars locally to give competition to their US and European counterparts
 Solution: Designed a multiple linear regression model for the prediction of car prices
 Key Achievement: Created a model with an MSE and R2 Square value of 81%

Domain: Telecom Industry | Tech Stack: Python, Jupyter Notebook | Aug 2019
 Objective: To reduce customer churn, telecom companies need to predict which customers are at high risk of churn
 Solution: Created SVM (Default)-linear, SVM (Default)-rbf, SVM(rfb ) [Hyper], RandomForest (Default),RandomForest
(Hyper), XGBoost (Default), XGBoost (Hyper Tuned) to predict churn data for future dataset or production.
 Key Achievement: Created a best predictive model using SVM and Random forest to predict the churn
Domain: Modifying Viterbi algorithm| Tech Stack: Python, Jupyter Notebook | Sep 2019
 Objective: Need to modify the Viterbi algorithm to solve the problem of unknown words.
 Solution: Modified Viterbi algorithm for unknown words.
 Key Achievement: build a modified Viterbi POS tagger algorithm is evaluated on the validation and the test datasets and
the results show some non-trivial improvement over the original algorithm.
Domain: Restaurant Search Chatbot | Tech Stack: Python, Jupyter Notebook | Oct 2019
 Objective: The bot will be able to 'talk' to users in English and will help them search for restaurants in several cities, of
multiple cuisine types, budgets etc.
 Solution: Designed NLP-Building Chatbots with Rasa to search restaurant in several cities.
 Key Achievement: build a chatbot and deploy it on a public channel such as Slack, Facebook etc.

Domain: Creating Classic NN| Tech Stack: Python, Jupyter Notebook | Nov 2019
 Objective: Predict the MNIST dataset.
 Solution: Create L-layered deep neural network and train it on the MNIST dataset.
 Key Achievement: build a Deep NN model with Test accuracy of ~87%.
Domain: Televisions | Tech Stack: OpenCV-Python, Python, Jupyter Notebook | Dec 2019
 Objective: To develop a cool feature in the smart-TV that can recognise five different gestures performed by the user
which will help users control the TV without using a remote
 Solution: Deployed a 3D Conv model that will be able to predict the 5 gestures correctly
 Key Achievement: Developed a model with a Conv3D of 72.47% Training 68% validation accuracy in 38th Epoch
Domain: Cab drivers| Tech Stack: Keras, Python, Jupyter Notebook | Feb 2020
 Objective: The goal of your project is to build an RL-based algorithm which can help cab drivers maximise their profits by
improving their decision-making process on the field.
 Solution: Deployed a Deep Reinforcement learning.
 Key Achievement: Developed a model with a DQN to optimize taxi driving strategies for profit maximization

Development Operations & Stakeholder Management
 Played a key role in formulating business strategies & managed development operations to accomplish top/bottom-line
 Identified and networked with prospective clients to generate business for increased profitability and growth.
 Ability to understand client requirements, design and drive the solutions. Design, understand and build computer vision
solutions leveraging deep learning and machine learning. Deep understanding of object detection, semantic and instance
segmentation, key point detection and object tracking algorithms. Experience in deploying deep learning solutions using
Docker. Ability to evaluate the latest research developments in machine-learning and deep-learning and help build state
of the art capabilities in these areas
Client Relationship Management
 Identified improvement areas & implemented Quality Control measures to maximize customer satisfaction levels
 Built and sustained healthy relationships with corporate clients by achieving timely delivery & quality control
 Delivered high quality services for customer delight & ensured optimum resource utilization for efficiency standards

Team Management & Leadership
 Spearheaded a team of ~30 to conceptualize and effectively implement the e-commerce sites for the company.
 Commissioned the development of DevOps Framework across all platforms in AWS.
 Building team with knowledge of AWS, AEM, Java, MySQL, Oracle, and RDS, web technologies like Servlet, JSP, Ajax,
HTML, JavaScript, CSS, UNIX Shell Script, Python, Hibernate 3.0, Web Sphere and JBoss, Elastic Search, Logstash and
Kibana Methodology for Data/log Analysis, Google Search Appliance, GitHub , Jenkins and Puppet for Automation

Project | AEM Digital and Ecommerce
 Supporting all the Digital website for Maintaining and enhancement of the functionality of the E-commerce sites as well
which has been developed through AEM
 End to End implementation in cloud with DevOps Framework

Engaged with client to gather information about existing infrastructure and designed architecture as per their
requirement, created share stack environment to serve different services (Jenkins, puppet, rundeck, database &
monitoring) on demand

 Involved in design, documentation, review, and closure. Created and customized workflows. Configured Jenkins server for
build deployment and created Jenkins jobs for each environment.
 Fixed bugs, resolved critical issues, code and design reviews, part of core development team Java/ J2EE and CMS.