Looking for a job

Sri Charan Rachakonda

@sricharanrachakonda

GenAI Engineer at Wells Fargo building Graph-RAG knowledge stores and agentic reporting workflows for financial documents.

United States

Message

At Wells Fargo, I engineered an AWS Textract document extraction pipeline that reduced manual data preparation by 40%. I also implemented a Neo4j Graph-RAG knowledge store for reasoning over complex entity relationships.

I developed financial report generation pipelines with LangChain and prompt engineering, and orchestrated agentic workflows with Microsoft AutoGen for data retrieval, validation, and report drafting. I adapted foundation models with LoRA and QLoRA fine-tuning for domain-specific terminology.

At Max Healthcare, I built machine learning models for fraud detection in claims, including BERT and LSTM models for clinical narratives and billing patterns. Earlier, at SBI Life Insurance, I developed churn prediction models and AWS Glue ETL pipelines.

Experience

Work history, roles, and key accomplishments

Current

GenAI Engineer

Current

Wells Fargo

May 2024 - Present (2 years 4 months)

Engineered an enterprise document extraction pipeline using AWS Textract and workflow automation, reducing manual data preparation by 40%. Implemented Graph-RAG knowledge stores and agentic workflows to automate compliance-driven reporting.

AWS Workflow Automation Neo4j Graph RAG Lang Chain Prompt Engineering AutoGen LoRAs QLoRA AWS Sagemaker Autoscaling Inference Endpoints Secure APIs Role Based Access Control Linux Logging Monitoring

Data Scientist (ML)

Max Healthcare

Jul 2022 - Jul 2023 (1 year)

Designed an ensemble fraud detection system using XGBoost and Isolation Forest to improve anomaly identification. Built BERT and LSTM models for clinical narrative analysis and deployed real-time risk scoring on Azure ML and AKS.

XGBoost Isolation Forest BERT LSTM SMOTE Cost Sensitive Learning Azure ML AKS Flask SHAP Fraud Detection Anomaly Detection

Data Scientist

SBI Life Insurance

Jul 2021 - May 2022 (10 months)

Developed Random Forest and Gradient Boosting models to predict customer churn with 82% accuracy. Built AWS Glue ETL pipelines processing over 10 million records monthly and designed A/B tests and Power BI dashboards to inform retention strategies.

Python Random Forest Gradient Boosting AWS Glue ETL A B Testing Power BI Customer Churn Prediction