Sai Sudeepthi

1) Email Analysis Automation and Optimization System - EAAS Tool (November 21 - Present)
Aim: Automate the process of IMS entry from the outlook to the database which will help in addressing and resolving the issues quickly. Outcome: Algorithm is able to extract the information from the outlook and pass it on to issue resolution in minutes where the turn around time in traditional process used to be 2 to 3 working days. Includes savings of $200K per year.

Technologies used: Text Mining in Python, SQL.

2) Collections Forecast | Developer (March 20 - Present)

Aim: Building Forecasting model that forecasts the amount to get collected for the month at day wise.

Outcome: Average accuracy of 98% for the year. Tools and Technologies used: Time Series in R, SQL.

3) MTV To CAS Transfers | Developer (May 21 - Present)

Aim: Transfer the claims from MTV platform to CAS platform in order to get collected. This will increase the collection process. Outcome: Collections of $5M was from the model in the year 2021 and $2M in the year 2022 (till date).

Tools and Technologies used: Linear Optimization model and Survival Analysis in R, SQL.

4) Write off's Decision Tree | Developer (June 21 - Present)

Aim: Predict the Claims that got to write off's and help the finance team in preventing them going to write off's.

Outcome: Accuracy in the range 95% - 98%. Helps in savings of $50K/month

Tools and Technologies used: Decision Tree, XGBoost in R, SQL

5) Vendor Finding letters | Developer (Aug 19 - March 20)

Aim: Classify if a claim is overpaid or not with the help of letters sent to the vendors.

Outcome: Model built with accuracy of 70%-75%

Tools and Technologies used: Text mining , Random Forest, XGBoost in python.