Navya Mohan Email & Phone Number

Irving, Tx, Us

• Onsked Scheduler – Developed and implemented a schedule management solution on Azure Machine Learning, streamlining appointment tracking for individuals, families, and businesses.• Gen AI Synthetic Data Generation – Led synthetic data generation using AWS Glue for data integration and transformation, AWS S3 for storing datasets, and AWS SageMaker for model training, ensuring data quality through comprehensive evaluation.• Feature Engineering for Customer Data – Spearheaded customer data feature engineering, enhancing data quality and relevance through preprocessing, EDA, and feature extraction with AWS Glue and AWS S3.• Machine Learning Model Implementation – Led the implementation and deployment of ML models on AWS SageMaker for model hosting, AWS EC2 for scalable computing, and AWS S3 for model artifacts storage, ensuring seamless integration for real-time decision-making.• Chatbot for Leave Retrieval – Implemented a chatbot for leave retrieval using RAG architecture and vector database for efficient document handling and retrieval.• Document Chunking and Embeddings – Processed documents using chunking methods and converted them to embeddings with OpenAI embeddings for enhanced information retrieval.• LLM-Based Natural Language Understanding • OpenAI Integration for Leave Management • Data Pipeline Optimization with Python• Automated Testing and Monitoring – Created automated testing and monitoring scripts in Python to validate model performance and track key metrics, enhancing model reliability and operational efficiency.Environment: Python, AWS SageMaker, Bedrock, AWS EC2, AWS ECR, AWS S3, AWS Glue, Athena, Terraform, Azure ML, MLOps, vector database, Text2SQL, RAG, OpenAI, NLP, SDV, Power BI

India

A banking domain-based project. The aim of the project is to create synthetic data for the bank customer database due to security reasons. Roles & Responsibilities:• Pre and post processing of data –Automate the preprocessing of data for data cleaning and based on user specific requirements. Post processing of data to revert the custom changes.• Data Analysis and Parsing – Detailed data analysis for interpretation of categorical and numerical data characteristics. Meta data creation based on data characteristics.• Synthetic data creation – Collaborate with Datacebo for synthetic data creation. Use synthetic data vault library for evaluating synthetic data.• Metrics and performance evaluation – Perform tests such as Chi-squared and KS metrics for performance comparison between synthetic data and real data.Environment: Python, Azure Data Bricks, Spark, Pyspark, SQL, Data Analytics Platform, SDV

Thiruvananthapuram, Kerala, India

• SOTA automation process - Development and testing for vehicle in loop (VIL)o Noise Factor Development and Testing (NFT)• In vehicle experience improvement using machine learningo Data acquisition, Data preparation, Feature extractiono Machine learning algorithm implemenation - based on the random forest and SVM.• Data logger configuration (both in bench and remotely) and linking logger to server.o Script creation for loggers based on vehicle type and model year.o Installation of data logger in vehicle. Perform different verification checks such as pre installation checks on the bench, checks with installation on a test vehicle, tests after deployment of logger.• Investigation on Heart Rate Variability for Driver Stress Analysiso Problem Analysis & Identification, Information gathering, Database collection.o Design, Matlab Coding, Debugging, Implementation and Testing• Self-learning car projecto Feature feasibility study, detailed literature survey and benchmarkingo Feature related signal identification. Matlab coding for analyzing data.o Pattern recognition and machine learning to find similar patterns and repeating scenarios.

Master Of Technology - Mtech, Digital Image Computing, 88%

Bachelor Of Technology - Btech, Computer Science And Engineering, 76.67

Bachelor'S Degree