Harshavardhan Naidu Modepalli

About Harshavardhan Naidu Modepalli

I am a GenAI Engineer and AI/ML Engineer with nearly 9+ years of experience designing, building, and deploying enterprise-scale AI and data-driven systems across highly regulated and data-intensive industries, including financial services, retail, insurance, and healthcare. Currently at Goldman Sachs, I engineer secure, production-grade LLM platforms and internal AI assistants that enable front-office, risk, and compliance teams to interact with proprietary data using natural language, while meeting strict security, governance, and regulatory requirements. My core expertise lies in Retrieval-Augmented Generation (RAG), LLM orchestration, and end-to-end AI system design, integrating structured and unstructured data sources to deliver accurate, explainable, and scalable AI solutions. I partner closely with data engineering, platform, and business teams to ensure reliability, performance, and real-world usability. Across my career, I’ve consistently delivered AI solutions that improve decision-making, reduce manual effort, and unlock actionable insights, leveraging modern cloud platforms (AWS, Azure, GCP) and production-ready ML engineering practices.

Experience

1. Gen Ai & Senior Data Engineer & Marketing Analytics Leader

   DICK'S Sporting Goods

   Jan 2025 — Present · US

   Architected enterprise-scale data and AI platforms using Databricks Lakehouse architecture with Apache Spark, PySpark, and Delta Lake to support large-scale analytics and machine learning workloads. Designed optimized ETL/ELT pipelines for structured and semi-structured datasets, ensuring reliable ingestion, schema enforcement, and high-performance query execution.• Built scalable batch and real-time data pipelines using Apache Spark, Kafka, and Airflow to process high-volume enterprise datasets. Implemented data validation, normalization, and feature engineering pipelines to prepare AI-ready datasets for downstream analytics and ML systems.• Designed advanced Retrieval Augmented Generation (RAG) architectures integrating LangChain, vector search, hybrid retrieval, and contextual re-ranking for enterprise knowledge intelligence systems. Leveraged LLMs such as GPT-4, Claude, Gemini, and Llama to enable domain-aware AI query processing and document intelligence.• Developed embedding generation pipelines using BERT-based models and foundation embeddings to enable semantic search and intelligent document retrieval. Integrated vector databases such as Pinecone and FAISS to support scalable similarity search and contextual knowledge retrieval.• Engineered intelligent data platforms integrating Snowflake, Snowflake Cortex, and PostgreSQL to combine structured and unstructured data processing. Built backend microservices using Python and FastAPI with asynchronous programming to expose high-throughput APIs for AI inference and data services.• Developed full-stack AI applications using ReactJS and Python APIs to deliver interactive dashboards and real-time operational insights. Enabled business users to interact with AI-powered analytics and knowledge systems through intuitive web interfaces.• Architected scalable cloud infrastructure on AWS (Bedrock, EKS, EC2, S3, Lambda, RDS, IAM, VPC) and Google Vertex AI for secure AI deployments.

Education

• Anna University Chennai

  Bachelor of Science, Computer Science

• Northeastern University

  Master of Science, Information Technology

  2025