SHUBHAM YADAV

Collection Models (Predictive Modeling and Deployment with Azure)-

As part of an internal tool, developed machine learning models to predict customer personas like high risk and low risk for unpaid bills. Subsequently, used MLops (Azure Devops and Azure services) to create a tool that enables clients to upload their customer data regularly and receive individualized predictions. These predictions empower clients to employ services like automated calls and messages for effective reminders about unpaid bills.

Credit Card Bills –

• Developed customer persona prediction machine learning models, covering the entire ML lifecycle from data collection to testing and model finalizing.
• Developed a Flask app that processes raw data individually to predict customer personas and established a CI/CD pipeline to build a Docker image for deploying an API.
• Executed the entire tool flow using Azure Function App, taking inputs from the user interface, and storing data in the database.

Internet Service Bills-

• Developed customer machine learning models to predict contact priority based on the probability of customer payment.
• Created a Flask app for bulk raw data processing to predict customer personas and implemented a CI/CD pipeline for Docker image creation, facilitating API deployment. Additionally, executed the entire tool flow via Azure Function App.

SKILLS – Machine Learning, Statistics, Azure Devops, Azure, Docker, Flask

Interaction Insights of customer (NLP using generative ai(chatgpt) and Azure)

• Created a tool capable of extracting insights from text data in six different languages, specifically focusing on chat conversations between customer service agents and customers.
• The tool categorized the text data into four key categories - Agent, Process, Service, and Technology - and predicted customer loyalty, churn, journey stage, effort, and the intent behind customer contacts using the ChatGPT API for advanced language processing.
• Deployed the tool on Azure using a Function App for efficient and scalable processing of the data, enabling seamless integration into the existing infrastructure.

SKILLS – NLP(Natural Language Processing), Azure, OpenAI, Mysql

News Classification

• Gathered extensive news data from the archives of news websites dating back two decades using selenium.
• Developed news classification models using Transformer-based ML models like BERT and ALBERT from the Hugging Face library.

Data Scraper

• Developed a web scraper tool utilizing Selenium, BeautifulSoup, and requests to efficiently gather reviews and other information from e-commerce websites.
• Integrated a user-friendly UI with Flask, drastically reducing manual work from 15 minutes to just 1 minute for the same tasks.

SKILLS - LLM models, Selenium, Deep learning