Nandhini Balakrishnan

Data Scientist at EmpowerMX
  • Claim this Profile
Contact Information
Location
McKinney, Texas, United States, US
Languages
  • English -
  • Malayalam -
  • Tamil -

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

You need to have a working account to view this content.
You need to have a working account to view this content.

Credentials

  • Google Data Analytics Certificate
    Coursera
    Apr, 2023
    - Sep, 2024
  • Advanced Google Analytics
    Google
    May, 2021
    - Sep, 2024
  • Microsoft Certified: Azure Fundamentals
    Microsoft
    May, 2021
    - Sep, 2024
  • Python Basic
    HackerRank
    Mar, 2021
    - Sep, 2024
  • Industrialized AI Data Engineer
    DXC Technology
    Feb, 2021
    - Sep, 2024
  • Industrialized AI Data Scientist
    DXC Technology
    Feb, 2021
    - Sep, 2024

Experience

    • United States
    • Aviation and Aerospace Component Manufacturing
    • 1 - 100 Employee
    • Data Scientist
      • Apr 2023 - Present
    • United States
    • Hospitals and Health Care
    • 700 & Above Employee
    • Data Scientist
      • Oct 2022 - Feb 2023

      ●Preprocessed and aggregated a million records and performed a predictive analysis that helped to reduce the manual labor associated with claim reviews. ● Performed data wrangling and ingested the model into the cloud-based AI tool (Aible) and defined the cost for every wrong prediction and every correct prediction by using the confusion matrix. ● Applied per unit labor(investigation) costs for each of the denied claims, and trained the model with classification ML algorithms thereby producing optimized value ● Achieved an accuracy of 83% with the best fitting model as XGB boost by evaluating the investigation efforts after running the model predictions. Show less

    • United States
    • IT Services and IT Consulting
    • 700 & Above Employee
    • Data Scientist
      • Aug 2021 - Feb 2023

      Developed a Flask web application utilizing appseed template that interacts with model predictions to submit, classify, and view the claim records ● Performed EDA on data, utilized SMOTE, Hash Encodings, Truncated SVD methods for feature selection & dimensionality reduction. ● Implemented a hypothesis and evaluated the data with different ML models by using Logistic models and autoviml with an accuracy score of 94% and f1 score of 80%. MLOps Accelerator - implemented and utilized the DXC MLOps Accelerator tool to deploy machine learning models in the Azure cloud for clients. ● Created Terraform scripts to provide infrastructure and cloud resources for various settings. ● Developed a pipeline for data preparation to register the dataset and ML datastore. ● Built a CI/CD pipeline in Azure DevOps to train the machine Learning model, to register and deliver to Kubernetes. ● Utilizing Azure Machine Learning Service to produce reusable Model Pipelines for model training, evaluation, and monitoring. ● Developed and maintained an Azure Dashboard to track the effectiveness and application of machine learning models. AI -starter – Web development ● Developed front end website architecture using react , MUI and other UI components, including translating designer mock-ups and converting wireframes into frontend code ● Designed user interaction on web pages and ensure cross platform optimization and browser compatibility ● Developed API’s using flask framework and python libraries. ● Performed Code review and merge pull request, ensured the code changes are deployed on every sprint demo. ● Created the Application code pipeline for test, dev and prod environments. Show less

    • Data Scientist
      • Sep 2019 - Jun 2021

      VA Artificial Intelligence Tech sprint (Health care & Innovation) -Suicide Prevention Chatbot Development● Working on Industrialized AI boot camp centered to deliver a Minimum Viable Product with a 12-week engagement.● Developed a pipeline to access the client data from the Azure data lake and retained the raw data bias-free and converted it into training-ready data for the bot.● Cleaned & Preprocessed 4 GB of client data that was in text form using NLP techniques like Lemmatizer, Stemmer, Regex.● Created a new derived variable risk co-efficient using regression model from the available target variable in the dataset to tag the illnesses with different levels of severity for suicide.● Developed user persona for the chatbot to classify patients with particular risk groups based on suicide attempts & illness.● Acquired good, bad, and harmful utterances datasets from different sources and performed NLP techniques like contractions, entity recognition filter, and sequence matcher to increase the quality of acquired data being fed to train the bot for intent detection. Show less

    • Infra setup and Data modelling
      • May 2020 - Dec 2020

      A web application to automate Airbus Quote on layover & Predict the Defect occurrences.● Designed application architecture based on client requirements and proposed the tech stack to kick start the application.● Created Data Flow Diagram (DFD) which served as a reference point for the application flow.● Created DB schema & wrote queries utilizing the DFD. Optimized the queries to reduce the response time by 30%.● Handled data ingestion on Azure blob, Azure SQL, Azure DSVM and configured the firewall rules to secure the Azure blob by Whitelist the incoming IPs.● Performed Exploratory Data Analysis (EDA) using Matplotlib and Plotly. ● Utilized Azure ML studio & DSVM for developing a predictive model by employing a logistic regression algorithm to estimate the airplane defects using historical data that reduced the manual effort and achieved model accuracy of 93%.● Aligned architecture and proposed a new ML ops architecture based on client requirements. Show less

    • Data Scientist
      • Sep 2019 - Jun 2020

      An iOS Mobile application to recommend food to the user-centered on user interest and lifestyle. ● Followed Agile Scrum methodology and was part of core AI Team. Contributed ideas on sprint tasks & stories in sprint planning meetings.● Built & tested a Hybrid recommender that works on Collaborative Filtering and SVD algorithms.● Designed a complex formula to avoid recommender cold start problems by rating the dishes as per user profile and USDA health guidelines. ● Optimized the code using Lambda, Map functions, and list comprehensions thereby minimized code redundancy and improved the performance of the code by 40%● Developed Azure function to expose the Recommender system as an endpoint that could be consumed by Frontend and obtain the response directly. Show less

    • Junior Data Scientist
      • Jul 2018 - Aug 2019

      Created data models and built pipelines for data processing in azure data bricks and conducted exploratory data analysisusing Plotly and seaborn.● Implemented the ML models in sales prediction using gradient boost and linear regression techniques.● Utilized Tableau dashboards to analyze as well as to visualize the preprocessed data. Generated 15 interactive reports usingTableau.

    • Application Developer
      • Nov 2016 - Jun 2018

      ● Created a web app that triggers backup action plans for employees and dry ran a Business Continuity plan using web app.● Developed user interface using JS, CSS, HTML backend services in C# MVC.● Created Stored Procedures and optimized SQL queries that helped in reducing the overall latency of the application.● Monitored incoming traffic & Investigated production server-related issues using the Splunk tool and assisted to resolve the high-priority security incidents.● Organized code reviews and collaborated with different teams for bug fixing. Show less

Education

  • Drexel University
    Master of Science - MS, Data Science
    2019 - 2021
  • Sri Ramakrishna Engineering College
    Bachelor of Engineering - BE, Electronics & Instrumentation Engineering
    2012 - 2016
  • Bharatiya vidhya bhavan
    Higher Secondary, Computer Science
    2010 - 2012

Community

You need to have a working account to view this content. Click here to join now