Hema Sagar Endluri

• at AGS Health developed Logistic regression machine learning model to predict a successful appeal on first time denied application.
• I worked on audio files; each file has almost 5 to 10 minutes. need to convert into text.
• To make them text, i used open-source modules Whisper from Openai. its helpe me alot compared to other models
• Agent and provider will be in each call, i should summarize their conversation using LLM model with fine-tuning.

• To maximize crop production using genomic and soil samples data, worked on various files and models.
• Worked on different data formats such as text, CSV, JSON, VCF, TXT, bed, bim, fam and performed machine learning algorithms in Python.
• Performed EDA and Descriptive analystics. Experimented with various predictive models including Logistic Regression, Support Vector Machine (SVM), Random Forest, XGBoost, Decision Tree, H2o to check for the best accuracy
• Performed Feature Engineering to generate many useful features.
• Developed Feature Selection framework.
• Model building framework for trying various Models like Linear Regression, Rigde, Lasso, KNNRegressor, RandomForestRegressor, GradientBoostingRegressor, XgboostRegressor.Automated repetitive tasks using scripting languages such as Python , saving time during the analytical process significantly.
• Deploy Machine Learning Architecture on AWS SageMaker
• Deploy the full machine learning architecture on AWS Sagemaker for model building, retraining, optimization and deployment
• Dockerized the Steady State Extraction Framework, Modelling building Framework, Feature Selection Framework, Optimization Framework and Deploy in ECR(Elastic Container Registry)
• Create Lambda(serverless) functions to invoke these algorithms from SageMaker

Deploy all the frameworks as REST API