Yamini Akula

Experience

• 

  AICTE

  Jul 2020 - Aug 2020

  Machine Learning Intern
• 

  Solenis

  Jan 2021 - Aug 2022

  Associate Business Analyst

  FP&A and revenue team•Created and integrated ETL packages to engineer data and code into Tableau dashboards monitoring 150M revenue, which I successfully owned and maintained, continually collaborating with clients, and providing insights to over 100,000 customers to ensure efficient and impactful data visualization.• Achieved a 90% data accuracy rate within a 2-month timeframe to streamline a data migration project using Azure Data Factory and SQL Server Integration Services (SSIS) and was recognized with an award for customer focus for delivering exceptional results under tight deadlines and increasing customer satisfaction score by 40%.• Orchestrated and automated seamless data migration using ETL technologies (Azure Data Factory, SSIS) through the development of stored procedures and SQL scripts for large-scale relational database entities across Enterprise Data Warehouses, reducing data processing time by 60%. Show less

• 

  Deloitte

  Aug 2022 - Jan 2024

  Data Scientist

  Audit and Assurance Learning analytics team•Innovated and deployed a RAG-based application using a Deloitte-approved ChatGPT model, integrated into the learning platform to automate the generation of questions, answers, and essays. This solution optimized user engagement, reduced SME intervention by 60%, and increased the retention rate by 30%.• Employed analytical skills utilizing collaborative filtering algorithms and Tableau to model a recommendation system, closely monitoring, and analyzing user interaction and engagement metrics on a learning platform for 80,000 users. This approach informed strategic recommendations, increased user engagement by 30%, and earned a spot award for project success.• Deployed a sentiment analysis system for MS SharePoint comments using the Logistic Regression algorithm in Python and scikit-learn, along with Word2Vec and BERT. Leveraged NLTK for text preprocessing. Operationalized the model via Tableau and Python scripts, achieving 85% accuracy and simplifying complex insights for technical and non-technical stakeholders.• Utilized Python (Pandas, NumPy) for data cleaning, processing, and analysis, reducing manual review hours from 100 to 8, optimizing overall efficiency.• Conducted A/B testing and hypothesis testing to validate the effectiveness of the sentiment analysis model, achieving an F1- score of 0.82. Show less

• 

  University of Massachusetts Dartmouth

  Oct 2024 - now

  Data and reporting analyst