Generative AI Engineer with Multi-Agent Mastery

Generative AI Engineer

Rithwika Narla

+1-980-***-**** ****************@*****.*** https://www.linkedin.com/in/rithwikanarla/ PROFESSIONAL SUMMARY:

Generative AI Engineer with 5+ years of experience designing and deploying LLM-powered AI/ML solutions and advanced Agentic AI systems across enterprise environments. Expertise in building multi-agent architectures using LangGraph and LangChain Agents for autonomous task planning, tool orchestration, memory management, and multi-step reasoning. Strong specialization in Prompt Engineering, LoRA/PEFT/QLoRA fine-tuning, Inference Optimization, and Retrieval- Augmented Generation (RAG). Hands-on experience with OpenAI APIs, Azure OpenAI, AWS Bedrock, Vertex AI, Hugging Face Transformers, GPT-4, and LLaMA 3. Designed AI Orchestrator Agents integrating enterprise systems (CRM, EHR, Billing APIs) with dynamic Tool-Augmented LLM Agents leveraging SQL tools, REST APIs, and Vector Databases (FAISS, Pinecone, ChromaDB). Proficient in Python, SQL, and scalable deployment using FastAPI, Docker, Kubernetes, CI/CD, with strong foundations in Deep Learning (CNNs, RNNs, LSTMs, Transformers), LLMOps, and cloud platforms (AWS, Azure) to deliver secure, production-grade GenAI and Agentic AI solutions. TECHNICAL SKILLS:

Programming Languages: Python (NumPy, Pandas, SciPy, Seaborn, Beautiful Soup, Scikit-learn, TensorFlow, PyTorch, NLTK, Hugging Face), R, SQL, JavaScript (Node.js, React.js), Shell Scripting.

Machine Learning & AI: LLM Chains, OpenAI API, Azure OpenAI Studio, AWS Bedrock, Hugging Face Transformers, Retrieval-Augmented Generation (RAG), Vector Databases (FAISS, Pinecone, ChromaDB), LoRA/QLoRA, Prompt Engineering, Fine-tuning LLMs, MLflow, Databricks, Deep Learning (CNNs, RNNs, LSTMs, Transformers, GANs).

Cloud & MLOps: AWS (S3, EC2, Lambda, SageMaker, API Gateway, DynamoDB, Bedrock), Azure (ML Studio, Cognitive Services, OpenAI Studio, Data Factory), Google Cloud (BigQuery, Vertex AI, Cloud ML), Docker, Kubernetes (EKS, AKS, GKE), Terraform, CI/CD Pipelines (Jenkins, GitHub Actions, GitOps).

Big Data & Distributed Computing: Apache Spark (PySpark, SparkSQL, MLlib), Hadoop Ecosystem (HDFS, MapReduce, Hive, HBase), Apache Airflow, Kafka, Snowflake, Data Lakehouse.

NLP & LLMs: GPT-4, BERT, T5, LLaMA, Stable Diffusion, LangChain, SpaCy, NLTK, Hugging Face, Intent Recognition, Sentiment Analysis, Named Entity Recognition (NER), Speech-to-Text.

Databases: SQL (PostgreSQL, MySQL, Oracle, SQL Server), NoSQL (MongoDB, Cassandra, Neo4j, DynamoDB, Redis).

Data Visualization & Analytics: Tableau, Power BI, Matplotlib, Plotly, ggplot2, Looker, Google Data Studio, Lakehouse Monitoring.

AI Security & Compliance: Explainable AI (XAI, SHAP, LIME), Model Governance, GDPR, HIPAA, SOC2 Compliance, AI Bias

& Fairness.

Version Control & Collaboration: Git, GitHub, GitLab, Bitbucket, JIRA, Confluence. PROFESSIONAL EXPERIENCE:

Client: Graham Healthcare Group, Troy, MI. Aug 2024 - Present Role: Generative AI Engineer

Responsibilities:

Designed and developed Conversational AI solutions, integrating LLMs (GPT-4, BERT, T5) with Azure OpenAI Studio, Microsoft Bot Framework (MBF) Bots, and CRM systems to enhance customer interaction and engagement.

Built an AI-driven chatbot to answer questions based on the information provided using LangChain, RAG (Retrieval- Augmented Generation), and LLaMA 3 as the base model.

Proficient in building applications with OpenAI (GPT-4, DALL·E), LLaMA, Claude, and other foundation models using APIs and open-source libraries (Hugging Face, LangChain).

Designed and deployed LLM-powered call summarization and agent-assist solutions, improving contact center efficiency and response accuracy.

Designed a HIPAA-compliant Agentic AI orchestration framework integrating GPT-4 and LLaMA 3 with EHR, CRM, and billing systems for autonomous task execution.

Built Retrieval-Augmented Generation (RAG) pipelines using FAISS and ChromaDB to enable semantic search across enterprise knowledge bases.

Developed end-to-end ML pipelines for NLP, computer vision, and time-series analytics using Python, TensorFlow, PyTorch, and Hugging Face.

Engineered scalable ETL/ELT pipelines using Spark, PySpark, Kafka, Delta Lake, and Azure Databricks, processing millions of records daily.

Created secure ML inference APIs using FastAPI and gRPC for real-time and batch predictions.

Containerized and deployed ML services on Azure Kubernetes Service (AKS) with auto-scaling and high availability.

Implemented model monitoring, drift detection, and alerting using Azure Monitor and Application Insights.

Collaborated with data scientists, product owners, and MLOps teams in Agile/Scrum environments.

Experienced in data manipulation and analysis with Python libraries like Pandas, NumPy, and SciPy.

Skilled in data visualization using Matplotlib and Seaborn to communicate insights effectively.

Developed RAG-enabled Agentic AI solutions to retrieve account history, service logs, and policy documents for context- aware response generation.

Proficient in Auto-sklearn, an automated machine learning (AutoML) library for Python, streamlining model selection and hyperparameter tuning.

Tested machine learning and NLP models applying algorithms based on decision trees, logistic regression, and neural CNNs, RNNs, LSTMs, GANs, networks.

Worked on natural language processing for documentation classification, text processing using NLTK, SpaCy, TextBlob to find the sensitive information in the electronically stored documents and text summarization.

Developed machine learning models using recurrent neural networks – LSTM for time series, predictive analytics.

Proficient in developing AI and machine learning solutions using Python.

Expertise in generative AI technologies, including GPT, Stable Diffusion, LangChain, and AI agents, with experience in fine-tuning LLMs and prompt engineering for enterprise solutions.

Strong background in big data processing using Apache Spark, Databricks, and Hadoop, integrating ML models with large-scale structured and unstructured datasets.

Skilled in implementing deep learning models with frameworks like TensorFlow and PyTorch.

Experienced in natural language processing (NLP) for text analysis and sentiment analysis.

Developed tool-augmented Agentic AI agents capable of dynamic API calling, SQL querying, and enterprise system integration.

Skilled in data preprocessing, feature engineering, and model evaluation for ML projects.

Used Kubernetes to deploy scale, load balance, and manage Docker containers with multiple namespace versions. Environment: Python, PyTorch, Scikit-learn, LangChain, OpenAI, Llama 3.1, GPT, ChromaDB, Apache Spark, Kafka, Azure, SageMaker, Azure ML, TensorFlow, BERT, Airflow, MongoDB, Cassandra, Tableau, Jenkins, Docker, Kubernetes, Streamlit, PCA, t-SNE.

Client: AT&T, Hyderabad, India Jan 2022 - Nov 2023 Role: AI Engineer

Responsibilities:

Developed and deployed LLM and GenAI models using AWS Bedrock, Azure OpenAI Studio, and Google Vertex AI, integrating them into enterprise workflows for AI-driven automation.

Preprocessed and transformed large-scale financial datasets using Pandas, NumPy, and TensorFlow, ensuring data quality and model compatibility.

Built scalable AI inference APIs using Flask, FastAPI, and TensorFlow Serving on AWS, enabling real-time and batch model predictions.

Integrated LoRA and QLoRA techniques into deep learning pipelines for efficient fine-tuning of transformer models like GPT, BERT, and T5, optimizing domain-specific model performance.

Integrated LoRA/QLoRA fine-tuned transformer models into Agentic AI workflows for domain-specific telecom reasoning.

Implemented Retrieval-Augmented Generation (RAG) to enhance chatbot accuracy and context-aware AI-driven responses in financial and customer service applications.

Designed end-to-end machine learning solutions for customer analytics, churn prediction, and credit risk modeling.

Implemented NLP pipelines for document classification, text analytics, and information extraction using transformer- based models.

Integrated ML predictions into Power BI dashboards for real-time business insights and KPI monitoring.

Designed and implemented MLOps pipelines leveraging Azure ML, AWS SageMaker, Kubernetes (EKS, AKS, GKE), and terraform, ensuring robust CI/CD workflows for AI deployments.

Built scalable ML deployment pipelines supporting automated AI decision workflows, forming the foundation for advanced Agentic AI systems.

Developed NLP models for Conversational AI chatbots, integrating them with Microsoft Bot Framework, LangChain, and Hugging Face Transformers for seamless customer interactions.

Integrated TensorFlow models with SQL databases (PostgreSQL, Cosmos DB, Oracle) for data retrieval, training pipeline automation, and predictive analytics.

Deployed LLMs and AI models on resource-constrained environments using TensorFlow Lite, ONNX Runtime, and quantized transformer models for efficient AI processing. Environment: Agile, RAG, Python, OpenAI Studio, Airflow, LoRA, QLoRA, TensorFlow, AWS Bedrock, MLflow, Databricks, Spark, PySpark, PyTorch, SQL, Airflow, Prompt, Zeppelin, Docker, Kubernetes, Tableau, GenAI, CI/CD, AWS, LLM, Apache Airflow.

Client: Tata Consultancy Services (TCS) - Hyderabad, India April 2020 - Dec 2021 Role: AI Engineer

Responsibilities:

Developed AI models for predictive analytics, leveraging TensorFlow, PyTorch, and Scikit-learn to enhance decision- making and automation.

Built time-series forecasting models (LSTM, ARIMA, Prophet) to analyze and predict trends based on historical data.

Designed and implemented Conversational AI chatbots using Microsoft Bot Framework, AWS Lex, and Dialogflow, improving user interactions through NLP.

Fine-tuned BERT and GPT-based models for intent recognition, named entity recognition (NER), and text classification tasks.

Integrated vector-based search with FAISS, Pinecone, and Elasticsearch to improve information retrieval and document intelligence.

Implemented Retrieval-Augmented Generation (RAG) techniques to enhance AI-driven Q&A and contextual response generation.

Developed multi-cloud AI pipelines with AWS SageMaker, Azure OpenAI Studio, and GCP Vertex AI, ensuring scalability and performance optimization.

Automated ML model deployment and monitoring using Jenkins, GitHub Actions, Terraform, and Kubernetes (EKS, AKS, GKE).

Designed feature engineering pipelines using Pandas, NumPy, and Snowflake, optimizing data preprocessing workflows.

Developed low-latency inference APIs using Flask, FastAPI, and AWS Lambda, enabling seamless AI model integration. Environment: Azure SQL, AI, Azure Data Factory, Azure Databricks, TensorFlow, Azure Data Lake, HDFS, Hive, Pig, Impala, MapReduce, Spark, Python, Kafka, Tableau, Teradata, Pentaho, Sqoop. EDUCATION:

University of North Carolina Master of Science in Computer and Information Science – May 2025 CGPA: 3.7/4.00

Contact this candidate