Hi, I’m Meghana Thotakuri, a Senior GenAI Engineer with 10+ years of hands-on experience delivering enterprise-scale ML and LLM-driven systems across healthcare, financial services, retail, and aviation. I design end-to-end GenAI architectures, secure inference pipelines, guardrails, and observability frameworks to empower scalable, compliant AI at enterprise scale. I specialize in RAG solutions, agentic AI applications, and LLM orchestration, with extensive MLOps/LLMOps, traffic engineering, and low-latency serving across AWS and Azure environments. I thrive on building measurable business impact through robust governance, explainability, and HITL-enabled workflows.

Meghana Thotakuri

Hi, I’m Meghana Thotakuri, a Senior GenAI Engineer with 10+ years of hands-on experience delivering enterprise-scale ML and LLM-driven systems across healthcare, financial services, retail, and aviation. I design end-to-end GenAI architectures, secure inference pipelines, guardrails, and observability frameworks to empower scalable, compliant AI at enterprise scale. I specialize in RAG solutions, agentic AI applications, and LLM orchestration, with extensive MLOps/LLMOps, traffic engineering, and low-latency serving across AWS and Azure environments. I thrive on building measurable business impact through robust governance, explainability, and HITL-enabled workflows.

Available to hire

Hi, I’m Meghana Thotakuri, a Senior GenAI Engineer with 10+ years of hands-on experience delivering enterprise-scale ML and LLM-driven systems across healthcare, financial services, retail, and aviation. I design end-to-end GenAI architectures, secure inference pipelines, guardrails, and observability frameworks to empower scalable, compliant AI at enterprise scale.

I specialize in RAG solutions, agentic AI applications, and LLM orchestration, with extensive MLOps/LLMOps, traffic engineering, and low-latency serving across AWS and Azure environments. I thrive on building measurable business impact through robust governance, explainability, and HITL-enabled workflows.

See more

Experience Level

Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Intermediate
Intermediate
See more

Work Experience

Senior GenAI/ML Engineer at UnitedHealth Group
March 1, 2024 - Present
Led enterprise GenAI governance architecture on AWS designing runtime flow (API Gateway → IAM → AI Guardrails → LLM Inference → Arize AI Observability) for HIPAA-compliant clinical AI systems. Built API Gateway security layers enforcing IAM authentication, rate-limiting, request queueing, and token throughput management for high-concurrency LLM inference workloads. Implemented AI security guardrails for prompt injection defense, PII/PHI redaction, and data leakage prevention using Presidio, RBAC controls, and audit logging frameworks. Designed RAG-based clinical intelligence systems using LangChain with hybrid retrieval pipelines leveraging FAISS and Amazon Kendra for grounded prior authorization workflows. Built LLM observability frameworks using Arize AI (Phoenix) and CloudWatch to track hallucinations, retrieval quality, model drift, and inference performance. Developed agentic AI workflows using LangChain and LangGraph orchestration with tool chaining and HITL validation fo
AI/ML Engineer (Generative AI) at U.S. Bank
September 1, 2022 - February 1, 2024
Designed and deployed AI/ML and early-stage GenAI services on Kubernetes (AKS) within a distributed inference architecture (API Gateway → service layer → inference endpoints → observability stack) for high-throughput financial systems. Built API Gateway and service integration layers on Azure enforcing IAM authentication, request throttling, distributed tracing, and root-cause debugging across Azure API Management, Redis cache, and inference endpoints. Built early-stage GenAI-powered document intelligence pipelines using LangChain and Azure OpenAI (GPT-3.5 Turbo) to extract structured insights from financial documents, contracts, and customer records for contextual search and workflow automation. Developed semantic anomaly detection systems using BERT embeddings, FAISS, and vector similarity search to identify complex fraud patterns beyond rule-based detection using contextual representation learning. Designed embedding-based semantic search and retrieval systems using Azure Cogn
Senior Data Scientist / ML Engineer at Abercrombie & Fitch
December 1, 2019 - August 1, 2022
Designed scalable data integration pipelines using Azure Synapse Analytics to consolidate retail sales, inventory, pricing, and clickstream data into centralized analytical datasets for ML model development. Built distributed data transformation workflows using Synapse Spark 3.1 to cleanse, normalize, and enrich large-scale retail datasets for downstream analytics and forecasting pipelines. Developed time-series demand forecasting models using SARIMA and Temporal Fusion Transformers (TFT) to predict SKU-level demand across products, stores, and regions. Engineered feature pipelines capturing pricing elasticity, seasonality, promotional impact, and customer behavior signals to improve forecasting performance and stability. Built BERT-based NLP pipelines using Hugging Face Transformers for product classification, attribute extraction, and automated catalog enrichment. Developed semantic search and recommendation systems using FAISS and embedding-based retrieval for similarity-driven prod
Data Scientist at American Airlines
September 1, 2017 - November 1, 2019
Designed scalable data ingestion pipelines using AWS Data Pipeline and AWS EMR to integrate flight operations, booking, loyalty, and revenue datasets into Amazon Redshift for centralized analytics. Built distributed data processing workflows using PySpark on AWS EMR to cleanse, transform, and standardize large-scale airline datasets for training and batch scoring pipelines. Developed demand forecasting models using ARIMA and Facebook Prophet to support route-level demand planning, seasonal forecasting, and capacity optimization decisions. Built revenue optimization models using XGBoost and LightGBM analyzing booking patterns, route demand elasticity, and pricing signals for dynamic fare optimization. Implemented anomaly detection systems using Isolation Forest and statistical methods to identify irregular patterns in delays, cancellations, and revenue deviations. Designed optimized data warehouse schemas using Amazon Redshift Spectrum and built automated batch scoring workflows using A
Data Analyst / Python Developer at ITC Infotech
June 1, 2015 - June 1, 2017
Developed data ingestion and transformation workflows using Apache Hive, Apache Pig, and Python to process enterprise datasets from multiple source systems into centralized analytical environments. Built ETL pipelines to cleanse, normalize, and validate structured business data; wrote Python scripts using Pandas and NumPy to automate repetitive data preparation tasks. Supported large-scale batch processing using Azure HDInsight and PySpark for enterprise analytical workloads across multiple client engagements. Developed customer segmentation models using K-Means clustering and built predictive models using Scikit-learn including Logistic Regression and Decision Trees for business forecasting. Designed and maintained Tableau dashboards to visualize operational KPIs, customer behavior trends, and performance metrics for business stakeholders. Wrote optimized SQL queries and managed PostgreSQL database operations for reporting, validation, and analytical workflows. Implemented data qualit

Education

Bachelor of Technology (B.tech) in Computer Science at Sathyabama Institute of Science and Technology, Tamil Nadu
January 11, 2030 - June 29, 2026

Qualifications

Add your qualifications or awards here.

Industry Experience

Healthcare, Financial Services, Retail, Transportation & Logistics, Software & Internet