Home

Sujitha C - Sr. AI/ML Engineer

[email protected]

Location: New York City, New York, USA

Relocation: Yes

Visa:

Resume file: Resume__Sujitha_1774357514042.docx Please check the file(s) for viruses. Files are checked manually and then made available for download.

Name: Sujitha Cherukuthota AI/ML Engineer PH: +1 (757) 696-4716. [email protected] LinkedIn: www.linkedin.com/in/sujitha-cherukuthota-4575ab60 Summary: Senior Generative AI and Machine Learning Engineer with 10+ years of experience designing scalable AI solutions using Python, Large Language Models (LLMs), and cloud-native architectures across healthcare, financial services, and retail domains. Extensive experience building Generative AI applications using LangChain, LangGraph, LlamaIndex, and AWS Bedrock, developing enterprise Retrieval-Augmented Generation (RAG) pipelines that enable contextual knowledge retrieval from large unstructured datasets. Strong expertise in Natural Language Processing (NLP) and HuggingFace Transformers, implementing document classification, semantic search, entity recognition, and automated summarization systems across enterprise knowledge platforms. Experienced in designing LLM-driven microservices architectures, integrating vector databases, embeddings, and REST APIs to deliver scalable AI platforms supporting intelligent search and automated knowledge discovery. Proficient in developing machine learning and deep learning models using Scikit-learn, XGBoost, PyTorch, and TensorFlow for predictive analytics, classification, regression, and anomaly detection across large-scale enterprise datasets. Hands-on experience implementing vector search architectures using Pinecone and embedding models to enable high-performance semantic retrieval systems across enterprise AI applications. Experienced in building data engineering pipelines using Apache Spark, PySpark, AWS Glue, and Pandas, enabling scalable data ingestion and feature preparation for machine learning and AI workloads. Strong background in deploying scalable AI platforms on AWS cloud infrastructure, including Amazon S3, Amazon Redshift, Amazon EKS, and AWS Bedrock, enabling reliable model deployment and distributed inference workloads. Skilled in implementing MLOps and LLMOps pipelines, including Docker, Kubernetes, Terraform, and CI/CD automation, enabling reproducible model deployment and scalable AI infrastructure management. Experienced in monitoring and optimizing production AI systems using Datadog, Prometheus, and Amazon CloudWatch, ensuring reliability, observability, and operational performance across enterprise AI platforms. Proven ability to collaborate within Agile/Scrum environments, partnering with product managers, data scientists, and engineering teams to deliver production-ready AI and Generative AI solutions. Technical Skills: Programming Languages: Python, SQL, Bash/Shell Scripting Generative AI & Large Language Models: Generative AI, Large Language Models (LLMs), GPT-4, Claude, AWS Bedrock, LangChain, LangGraph, LlamaIndex, CrewAI, Prompt Engineering, Retrieval-Augmented Generation (RAG) Natural Language Processing & LLM Frameworks: HuggingFace Transformers, Sentence Transformers, Natural Language Processing (NLP), Text Embeddings, Semantic Search, Hybrid Search Machine Learning & Deep Learning: Machine Learning, Predictive Modeling, Classification, Regression, Clustering, Feature Engineering, Scikit-learn, XGBoost, PyTorch, TensorFlow Vector Databases & Retrieval Systems: Vector Databases, Pinecone, Vector Search, Approximate Nearest Neighbor (ANN) Data Engineering & Big Data: ETL Pipelines, Data Processing, Apache Spark, PySpark, AWS Glue, Pandas, Hadoop, Hive, HDFS Cloud Platforms: Amazon Web Services (AWS), AWS Bedrock, Amazon S3, Amazon Redshift, Amazon EKS MLOps & Infrastructure: Docker, Kubernetes, Terraform, CI/CD Pipelines, GitHub Actions, Jenkins, MLflow Monitoring & Model Evaluation: Datadog, Amazon CloudWatch, Prometheus, Ragas, TruLens Backend Development & APIs: FastAPI, Flask, REST APIs, Microservices Architecture, Object-Oriented Programming (OOP) DevOps, Testing & Tools: Git, Linux, PyTest, ELK Stack, Grafana, Confluence Methodologies: Agile, Scrum, Software Development Life Cycle (SDLC) Professional Experience: HCA Healthcare Richmond, VA Senior Generative AI / LLM Engineer Feb 2024 Present Architected enterprise-scale Generative AI knowledge assistant platform using Python, LangChain, and AWS Bedrock, enabling healthcare analysts to retrieve contextual insights from large unstructured clinical datasets through advanced LLM-driven semantic search and retrieval pipelines. Designed scalable microservices architecture integrating LLM inference services, embedding pipelines, vector databases, and REST APIs, enabling modular deployment of enterprise Retrieval-Augmented Generation (RAG) services supporting healthcare analytics and knowledge discovery applications. Delivered enterprise GenAI capabilities within Agile/Scrum frameworks, collaborating with product owners, clinical informatics teams, and engineering teams to iteratively develop LLM-powered applications aligned with healthcare compliance and enterprise AI adoption initiatives. Built automated data ingestion pipelines integrating FHIR APIs, Amazon S3 healthcare data lakes, and structured enterprise data sources, enabling ingestion of clinical documentation and operational datasets supporting downstream LLM-based retrieval systems. Implemented scalable data processing pipelines using Pandas, PySpark, and AWS Glue, transforming large healthcare document datasets into structured corpora optimized for embedding generation and RAG workflows. Designed scalable data lake architecture using Amazon S3 and Amazon Redshift, enabling high-performance storage and retrieval of enterprise healthcare datasets supporting LLM inference pipelines and Generative AI analytics workloads. Implemented enterprise vector database infrastructure using Pinecone, generating high-dimensional text embeddings and enabling fast Approximate Nearest Neighbor (ANN) vector search supporting large-scale semantic search and RAG pipelines. Integrated enterprise Large Language Models including GPT-4 and Claude via AWS Bedrock, enabling advanced reasoning capabilities supporting clinical document understanding, knowledge discovery, and automated summarization workflows. Developed scalable Retrieval-Augmented Generation (RAG) pipelines using LangChain and LlamaIndex, combining vector search, semantic retrieval, and contextual LLM reasoning to generate accurate responses grounded in healthcare documentation. Implemented semantic search and hybrid search architectures, combining embedding-based retrieval with keyword ranking techniques to improve contextual response relevance across large healthcare knowledge repositories. Built modular LLM orchestration services using LangChain, LangGraph, and REST APIs, enabling integration of document loaders, embedding models, and vector search services within scalable enterprise Generative AI microservices platforms. Applied HuggingFace Transformers components for embedding workflows and lightweight model experimentation, improving domain-specific healthcare terminology understanding within RAG systems. Developed Agentic AI orchestration workflows using CrewAI, enabling multi-step reasoning agents capable of autonomous document retrieval, summarization, and contextual knowledge generation across distributed healthcare data sources. Implemented parameter-efficient fine-tuning techniques including LoRA and PEFT, enabling lightweight domain adaptation of transformer models for healthcare terminology while maintaining efficient inference performance. Conducted enterprise LLM evaluation using Ragas and TruLens, measuring hallucination rates, grounding accuracy, and response relevance across production Generative AI knowledge retrieval pipelines. Containerized enterprise LLM inference services using Docker, packaging embedding pipelines and vector search services into reproducible environments supporting scalable deployment of Generative AI applications. Deployed distributed AI workloads using Kubernetes on Amazon EKS, orchestrating scalable microservices supporting embedding generation and LLM inference workloads. Implemented automated CI/CD pipelines using GitHub Actions, enabling continuous integration and controlled deployment of evolving Generative AI services. Managed infrastructure provisioning using Terraform Infrastructure-as-Code, automating deployment of AWS Bedrock integrations and vector database infrastructure. Implemented observability frameworks using Datadog and Amazon CloudWatch, monitoring LLM latency, token usage, and vector search performance across production AI systems. Developed automated testing frameworks using PyTest and produced technical documentation using Confluence to support cross-team knowledge transfer. Sallie Mae Newark, DE AI / Machine Learning Engineer May 2022 Jan 2024 Designed enterprise AI-powered document intelligence platform using Python, Natural Language Processing (NLP), and HuggingFace Transformers, enabling automated analysis and classification of financial documents and customer communications across enterprise banking workflows. Architected scalable machine learning microservices architecture integrating model training pipelines, REST API-based inference services, and feature engineering workflows supporting enterprise analytics and financial data intelligence platforms. Delivered AI capabilities within an Agile/Scrum development environment, collaborating with product managers, data scientists, and platform engineers to iteratively deploy production-ready AI and NLP solutions supporting enterprise decision-making. Developed automated data ingestion pipelines integrating REST APIs, enterprise Amazon S3 data lakes, and relational databases, enabling ingestion of structured financial datasets and unstructured customer communications used in machine learning pipelines. Implemented scalable data transformation pipelines using Pandas, PySpark, and AWS Glue, preparing large financial datasets for machine learning model training and NLP-based document analytics workflows. Designed scalable data lake architecture using Amazon S3 and Amazon Redshift, enabling secure storage and efficient retrieval of enterprise financial datasets supporting analytics, reporting, and AI model training workloads. Generated text embeddings using Sentence Transformers, enabling semantic representation of financial documents and early semantic search capabilities across enterprise knowledge bases. Developed advanced NLP models using HuggingFace Transformers, enabling automated document classification, named entity recognition, and financial document summarization across enterprise data platforms. Implemented early Retrieval-Augmented Generation (RAG) prototype systems, combining document embeddings, vector similarity search, and transformer-based language models to improve knowledge retrieval across internal banking documentation. Optimized machine learning models through feature engineering, hyperparameter tuning, and cross-validation, improving predictive performance across financial classification and document processing use cases. Built modular machine learning services using Python, FastAPI, and object-oriented programming (OOP) principles State of California San Francisco, CA Senior Data Scientist/ Machine Learning Engineer Feb 2020 Apr 2022 Led development of scalable machine learning platforms using Python, PyTorch, and TensorFlow, enabling state agencies to perform predictive modeling, classification, and regression analysis on large public datasets supporting operational planning and policy decision-making. Architected enterprise microservices-based machine learning architecture, integrating model training pipelines, feature engineering workflows, and REST API-based inference services supporting deployment of production machine learning models across multiple government analytics applications. Delivered machine learning capabilities within Agile/Scrum development environments, collaborating with policy analysts, data scientists, and engineering teams to iteratively develop and deploy enterprise predictive analytics solutions across government data intelligence platforms. Built automated data ingestion pipelines integrating REST APIs, structured government databases, and Amazon S3 data lakes, enabling ingestion of large public datasets used for machine learning training pipelines and advanced analytics workflows. Implemented distributed data processing pipelines using Pandas, PySpark, and AWS Glue, transforming large government datasets into structured feature datasets optimized for predictive modeling, classification, and clustering algorithms. Designed scalable data lake architecture using Amazon S3 and Amazon Redshift, enabling secure storage and high-performance retrieval of enterprise public datasets supporting machine learning training and analytics workloads. Developed advanced machine learning models using Scikit-learn, XGBoost, and PyTorch, enabling classification, regression, and anomaly detection across large government datasets supporting operational forecasting and analytics initiatives. Built deep learning models using TensorFlow and integrated HuggingFace Transformers libraries for advanced text classification and document analytics across large public administrative datasets. Implemented scalable feature engineering pipelines, extracting structured predictive features from complex datasets to improve model training accuracy and enable efficient machine learning experimentation workflows. Optimized machine learning performance through hyperparameter tuning, cross-validation, and model experimentation, improving predictive performance across multiple enterprise machine learning models. Implemented machine learning lifecycle management using MLflow, enabling experiment tracking, model versioning, and reproducible model training workflows across development and production environments. Containerized machine learning inference services using Docker, packaging trained models and feature engineering pipelines into reproducible environments supporting scalable machine learning deployment pipelines. Deployed distributed ML workloads using Kubernetes (Amazon EKS), enabling scalable model serving infrastructure supporting real-time inference workloads across enterprise analytics platforms. Implemented automated CI/CD pipelines using Jenkins, enabling continuous integration and controlled deployment of machine learning pipelines and model inference services. Implemented monitoring frameworks using Prometheus and Amazon CloudWatch, tracking model performance, data drift, and inference latency across production machine learning systems. Walmart Global Tech Bentonville, AR Machine Learning Engineer Oct 2016 Dec 2019 Developed scalable machine learning solutions using Python and Scikit-learn, enabling predictive modeling, classification, and regression analysis across large retail datasets supporting demand forecasting and customer behavior analytics for enterprise e-commerce platforms. Designed scalable microservices-based machine learning architecture, integrating model training workflows, feature engineering pipelines, and REST API-based inference services supporting production deployment of machine learning models across enterprise retail analytics applications. Delivered machine learning capabilities within Agile/Scrum development environments, collaborating with product managers, data engineers, and data scientists to develop enterprise predictive analytics solutions supporting customer intelligence and product recommendation systems. Built automated data ingestion pipelines integrating REST APIs, transactional databases, and enterprise Hadoop-based data lakes, enabling ingestion of high-volume retail datasets supporting downstream machine learning and analytics workflows. Implemented distributed data processing pipelines using Apache Spark and PySpark, transforming large-scale retail datasets into structured feature datasets optimized for predictive modeling, classification algorithms, and machine learning experimentation. Designed scalable data storage architecture using Hadoop HDFS and Hive, enabling efficient storage and retrieval of large historical retail datasets supporting enterprise machine learning training pipelines. Developed predictive machine learning models using Scikit-learn, implementing algorithms including Random Forest, Gradient Boosting, and Logistic Regression for customer segmentation and product demand forecasting. Implemented scalable feature engineering pipelines, extracting behavioral and transactional features from large retail datasets to improve predictive model accuracy and enable efficient machine learning training workflows. Built enterprise recommendation systems using collaborative filtering techniques, enabling personalized product recommendations across e-commerce platforms and improving customer product discovery experiences. Optimized model performance through hyperparameter tuning, cross-validation, and model experimentation, improving predictive performance across enterprise retail analytics applications. Containerized machine learning inference services using Docker, packaging model pipelines and feature engineering workflows into reproducible environments supporting scalable deployment of machine learning services. Implemented automated CI/CD pipelines using Jenkins, enabling continuous integration and deployment of machine learning models and analytics services across enterprise retail platforms. Developed monitoring dashboards using ELK Stack and Grafana, tracking model performance, inference latency, and system reliability across production machine learning systems. Citibank Hyderabad, India Data Engineer Aug 2015 Sep 2016 Developed enterprise data processing applications using Python and SQL, enabling automated extraction and transformation of high-volume financial transaction datasets used for operational reporting, analytics workflows, and internal risk analysis systems. Designed scalable data pipeline architecture integrating financial data processing modules, batch data workflows, and enterprise reporting services supporting internal banking analytics platforms and regulatory reporting systems. Worked within Agile/Scrum development environments, collaborating with business analysts, database administrators, and engineering teams to deliver reliable data processing applications supporting enterprise banking operations. Built automated data ingestion pipelines integrating enterprise relational databases, flat-file sources, and REST APIs, enabling ingestion of high-volume financial transaction data supporting analytics and reporting workflows. Implemented scalable ETL workflows using Python and SQL, transforming raw financial datasets into structured reporting datasets used for internal business intelligence dashboards and operational analytics systems. Developed scalable data transformation pipelines using Pandas, enabling efficient processing and validation of large financial transaction datasets supporting enterprise data analytics workflows. Implemented automated data validation and quality checks using Python, ensuring integrity and consistency of financial datasets across enterprise analytics systems and regulatory reporting pipelines. Built reusable Python modules using Object-Oriented Programming (OOP) principles, improving maintainability and scalability of enterprise data processing applications used across internal analytics teams. Optimized complex SQL queries and database interactions, improving performance of financial reporting workflows and enabling faster retrieval of large financial transactions. Developed lightweight REST APIs using Flask, enabling integration of Python-based data processing pipelines with enterprise reporting applications and internal analytics systems. Managed deployment of applications in Linux environments, ensuring reliable execution of enterprise data processing workflows across development and testing environments. Implemented automated testing frameworks using PyTest, validating ETL pipelines, data transformation logic, and enterprise reporting workflows. Produced detailed technical documentation using Confluence, documenting data pipelines, ETL processes, and system architectures supporting cross-team collaboration and knowledge transfer. Education: Bachelor of Technology (B.Tech), Sreyas Institute of Engineering and Technology, Hyderabad, India Keywords: continuous integration continuous deployment artificial intelligence machine learning sthree bay area Arkansas California Delaware Virginia