NIDHI MUDUGAL

• I work as a data engineer in the Claims Anomaly Detection product stream, which is essential for identifying and recovering instances of fraud, waste, and abuse in healthcare, significantly impacting financial, patient safety, and compliance areas.
• Working on the design, development, and maintenance of the data engineering pipeline, which plays a critical role in fraud prevention by analyzing billing patterns, identifying suspicious claims, and enabling swift action against anomalies using the modern technological platform, which comprises GCP, Snowflake, Exponential AI – ENSO, Python, and SQL.
• My team is responsible for the migration of 1,000+ legacy concepts to the Timber CAD platform. This involves leveraging various data-intensive pipelines during the analysis, development, testing, and deployment stages, and recently also scaling up the migration process by moving towards Python-based in-house analytics AI assist, which generates code using LLM. This helped to speed up the development effort of each concept migration from a few weeks to a couple of days.
• Feature enhancement and maintenance of the concept validation pipeline, which identifies data and coding errors by running on the OpenShift Container Platform powered by Kubernetes. The validation engine rules are written using Python and executed through the Bitbucket web hook.
• Work on R&D and automation to explore opportunities for innovation, like Analytic Redesign, where we created reusable rules to develop analytics by modularizing the design components. This reduced the analytic customization, improved the logging, and reduced the execution time by avoiding running redundant queries. Another use case was automating a process to identify if each recoverable claim in the universe matches the business requirement (aka hit rate), which saved a lot of UAT validation efforts by business users and improved the quality.
• Develop additional utilities to speed up the larger migration process by producing technical analysis to help decision-making, such as scraping the entire legacy UI platform to identify the concepts' complexity, code change frequency, reusability of objects, etc.
• Development of applications to process and generate invoices, manage Kafka consumer payloads, and handle logging configurations. It integrates with Snowflake for database operations and uses Kafka for message brokering. The project includes modules for reading configurations, logging, and consuming messages from Kafka topics. The primary goal is to automate the generation and management of invoices, ensuring efficient data processing and logging throughout the workflow.

• Identified and leveraged existing data to provide insight and assisted with data governance.
• Managed performance monitoring and tuning while identifying and repairing issues.
• Conducted data analysis to prepare forecasts and identify trends.
• Assisted solution providers with the definition and implementation of technical and business strategies.
• Collected, tracked, and reviewed data to evaluate business and market trends.
• worked on data cleansing, and data validation to ensure accurate data transformation and integration.
• Worked with the defect team to identify new defects and reported to the solution and leadership team.
• Ran quality controls on components and final products and maintained Excel spreadsheets.
• Developed and executed test plans and test script execution, reviewed new defects, wrote steps to test, outlined information for change requests, and notified clients of changes in end-user functionality
• Tested and deployed code, verifying operational and UX integrity for releases.
• Contributed to the QA team to perform testing meeting various parameters.
• Managed timely flow of business intelligence information to users.

• I gathered the requirements from the clients and automated the software with no user intervention, using PowerShell scripts.
• I organized a call with clients to understand the requirements of the business, and then we took over the software installation to ensure the software meets the business standards.
• I made sure the client systems are up-to-date by deploying software through the SCCM.
• Wrote, tested, and deployed code, verifying operational and UX integrity for releases.
• Involved in the QA team to perform testing, meeting various parameters.

Academic projects:

• Database Management for Album Collection Tool.

To create database views, a trigger, a stored procedure, and a stored function in MySQL to improve data retrieval and management in a music album collection database, with the aim of successfully implementing the project tasks, will improve the functionality of the music album collection database and facilitate the efficient retrieval and management of the data.

• Predicting the bond price from trading data.

Where data preprocessing, data cleaning, feature selection, and feature engineering techniques are involved, handling missing or incomplete data is essential. After that, evaluating different ML algorithms based on the data and task, and choosing an appropriate evaluation metric for the ML model is necessary. Model training involves splitting the dataset into training and validation sets, training the selected ML model on the preprocessed dataset, and evaluating model performance on the validation set. Finally, using the trained ML model to predict outcomes on the test set, evaluating the model's performance on the test set, and comparing the model's performance to the baseline or other models is crucial.