AI/ML Engineer - GenAI & RAG

Durga Prasad Gundepalli

AI/ML Engineer

*****@***********.*** 317-***-**** USA LinkedIn

Summary

AI/ML Engineer with 4+ years of experience building scalable Machine Learning and Generative AI solutions across BFSI, insurance, and telecom domains. Expertise in Python, PyTorch, TensorFlow, and LLM-based applications using LangChain, RAG, and Hugging Face. Strong hands-on experience in end-to-end ML pipelines, model deployment, and real-time inference systems on cloud platforms (AWS, Azure). Proven track record in delivering production-grade ML systems with measurable business impact. Technical Skills

• Programming & ML Frameworks: Python, R, SQL, Pandas, NumPy, Scikit-learn, XGBoost, LightGBM, TensorFlow, Keras, PyTorch, spaCy, NLTK, Generative AI, LLMs, LangChain, RAG, Hugging Face Transformers

• Cloud Platforms & MLOps: AWS (SageMaker, S3, EC2, Lambda), Azure (Azure ML, Databricks), GCP (Vertex AI), Docker, Kubernetes, MLflow, Airflow, CI/CD Pipelines, Flask/FastAPI, MLflow, CI/CD Tools, REST APIs, Microservices, Flask, FastAPI

• Data & Distributed Computing: Spark (PySpark), Hadoop (HDFS), Large-Scale Data Processing, Feature Stores, Snowflake, kafka

• Specialized ML Techniques: NLP (BERT, Transformers, NER, Text Classification), Time-Series Forecasting (ARIMA, Prophet, LSTM), Recommendation Systems, Causal Inference, A/B Testing Framework Design, End-to-end ML Pipeline, Hyperparameter tuning

• Visualization & Reporting: Power BI, Tableau, Matplotlib, Seaborn, Plotly

• Domain Expertise: Customer Churn Prediction, Fraud Detection, Credit Risk, Sales Forecasting, Intelligent Document Processing (IDP), Insurance & BFSI Analytics

Professional Experience

AI/ML Engineer, New York Life Insurance 10/2024 – Present Remote, USA

• Designed and implemented an end-to-end ML pipeline for customer churn prediction using XGBoost, including feature engineering, hyperparameter tuning, and model deployment, improving retention campaign efficiency by 20%.

• Built an optimized Claims Fraud Detection System leveraging Random Forest models and anomaly detection methods, significantly reducing fraudulent claim processing time by 25% while improving operational reliability and investigative decision-making efficiency.

• Developed an NLP + LLM-assisted document processing pipeline using BERT, OCR, and Generative AI techniques, enabling automated data extraction and improving processing efficiency by 40%.

• Integrated LLM APIs (OpenAI, Hugging Face) into production workflows to enable text generation, summarization, and contextual question-answering, improving processing efficiency and user interaction capabilities.

• Engineered scalable features and conducted extensive trend and seasonality analysis on multi-terabyte policyholder datasets using Pandas and Spark, substantially improving overall model input quality and downstream predictive performance consistency.

• Containerized machine learning models with Docker and deployed scalable, resilient inference services using AWS SageMaker and Kubernetes, ensuring continuous high availability, low-latency predictions, and seamless production integration.

• Established comprehensive model monitoring and automated retraining pipelines using MLflow, effectively maintaining performance drift below 2% and ensuring continuous regulatory compliance alongside robust lifecycle governance.

• Collaborated closely with cross-functional teams to integrate deployed ML solutions into Azure Data Lake and enterprise Kafka streaming infrastructures, enabling efficient real-time analytics and faster data-driven decision workflows. AI/ML Engineer, Infosys 01/2021 – 07/2023 Hyderabad, India Project 1: Customer Churn Prediction for Telecom Client

• Developed machine learning models integrating multi-armed bandit strategies and uplift modeling to optimize retention campaigns, reduce customer churn, and maximize revenue through targeted interventions based on predictive scoring.

• Conducted extensive exploratory data analysis and feature engineering on 10TB-scale telecom datasets using PySpark, Pandas, and SQL, identifying patterns, correlations, and key predictors of customer churn effectively.

• Built and tuned Random Forest, XGBoost, and Logistic Regression models, optimizing hyperparameters through cross-validation and grid search, ensuring high predictive accuracy, precision, and F1-score reliability.

• Developed scalable data pipelines using PySpark and SQL on large datasets, enabling efficient processing and model training

• Deployed production-ready predictive models via Flask APIs, enabling seamless CRM integration, real-time customer scoring, and actionable insights for proactive retention campaign execution. Project 2: Retail Sales Forecasting using Time-Series Analysis

• Developed advanced time-series forecasting models using ARIMA, Prophet, LSTM, and Transformer-based architectures, improving retail sales predictions and enhancing inventory planning and supply chain operational efficiency.

• Performed feature scaling, trend analysis, and seasonality decomposition on historical sales data, improving model accuracy while reducing forecast errors and providing actionable insights for business stakeholders.

• Built interactive dashboards and visualizations using Power BI and Matplotlib, allowing business teams to monitor sales trend s, forecast accuracy, and decision-making performance effectively.

• Implemented workflow orchestration with Airflow and utilized Azure Databricks for scalable distributed data processing, improving model retraining efficiency, scheduling, and automated reporting. Project 3: Intelligent Document Processing (IDP) for BFSI Client

• Developed NLP and OCR pipelines using BERT, Transformers, and Azure Form Recognizer to automatically extract structured data from unstructured financial documents, reducing manual processing time by 30%.

• Designed and deployed AI microservices on Kubernetes with CI/CD pipelines using GitHub Actions, ensuring scalable, maintainab le, and automated model deployment across multiple client environments.

• Explored LLM APIs and transformer-based models for text processing use cases, contributing to prototype development and model evaluation.

• Applied Named Entity Recognition and text classification techniques to accurately identify and extract critical information from various document formats, improving operational efficiency and reducing error rates.

• Monitored model performance continuously, retrained models periodically, and documented workflows, ensuring reproducibility, consistency, and seamless collaboration within Agile development sprints. ML Engineer, HSBC 01/2020 – 12/2020 Andhra Pradesh, India

• Developed Credit Risk Prediction Models using XGBoost, Random Forest, and Logistic Regression to evaluate borrower risk, improving loan approval accuracy and reducing non-performing assets.

• Applied SHAP values to interpret and explain model predictions for stakeholders, ensuring transparency, regulatory compliance, and improved trust in AI-driven credit decision processes.

• Applied SHAP for model interpretability within production ML systems to ensure transparency and compliance.

• Designed and implemented A/B testing frameworks to evaluate model deployment performance, validate predictive impact, and optimize credit scoring effectiveness across multiple customer segments.

• Created interactive dashboards with Tableau and Matplotlib to communicate high-risk segments, model outputs, and actionable insights to internal stakeholders, improving informed decision-making.

• Ensured regulatory and internal audit compliance at all stages of model development, maintaining strict adherence to data privacy, risk management, and documentation standards.

Education

Indiana University – Purdue University Indianapolis (IUPUI) Indianapolis, Indiana, USA Master of Science (M.S.), Applied Data Science 08/2023 – 05/2025 Sagi Rama Krishnam Raju Engineering College, JNTUK Andhra Pradesh, India Bachelor of Technology (B.Tech), Computer Science and Engineering 08/2018 – 08/2021 Smt. B. Seetha Polytechnic College Andhra Pradesh, India Diploma, Electronics and Communication Engineering 09/2015 – 04/2018

Contact this candidate