Home

Keerthi - Data Engineer, AI Engineer

[email protected]

Location: Addison, Pennsylvania, USA

Relocation:

Visa:

Keerthi D Executive Summary Accomplished Lead Data Engineer and Machine Learning Engineer with 12+ years of expertise in designing, developing, and deploying advanced AI, Machine Learning, and Generative AI solutions at an enterprise scale. Proven success in delivering predictive models, Retrieval-Augmented Generation (RAG) frameworks, and large language model (LLM)-based applications that drive strategic business insights, optimize operations, and enhance user experiences. Adept at building scalable data pipelines, architecting end-to-end MLOps processes, and leading cross-functional teams to deliver impactful, high-performance solutions. Core Competencies Machine Learning & AI: Generative AI, LLMs (RAG models), TensorFlow, Scikit-Learn, MLFlow, Spark MLlib Programming & Data Engineering: Python, SQL, PySpark, Scala, Java, C++ Big Data Technologies: Hadoop, Spark, Kafka, AWS Kinesis MLOps & CI/CD: Jenkins, Docker, Airflow, Git, AWS Step Functions Cloud Platforms: AWS (SageMaker, EC2, Lambda, S3), Databricks Visualization Tools: Tableau, Power BI, Matplotlib, ThoughtSpot Professional Experience Lead Data Engineer & Machine Learning Engineer Comcast | United States | Oct 2022 Present Developed an automated ticketing system leveraging LLMs (OpenAI GPT-4, AWS SageMaker), integrating with Microsoft Teams chatbot to streamline ticket categorization and resolution. Designed and implemented RAG models using Amazon OpenSearch and S3, achieving 95% accuracy in domain-specific ticket resolutions. Built predictive models for order forecasting, achieving over 90% accuracy, reducing inventory stockouts and saving millions annually. Engineered fraud detection algorithms using Databricks PySpark, reducing fraudulent activity by 30% through real-time monitoring and anomaly detection. Architected end-to-end MLOps pipelines with Jenkins, Docker, and Git, cutting model deployment cycles by 25%. Processed terabytes of data daily by developing high-throughput ETL pipelines with Kafka, Spark, and Hadoop, ensuring optimal data quality and reliability. Mentored a team of 6 engineers, fostering skill development and best practices, boosting team productivity by 15%. Data Transformation Manager Al Turki Enterprises | Oman | Apr 2017 Mar 2021 Led the development of ETL pipelines for HR and operational data, improving reporting accuracy by 25% and processing speeds by 30%. Built predictive models for inventory downtime prevention, achieving near-zero downtime and reducing operational costs significantly. Designed executive dashboards using Tableau, delivering actionable insights that enhanced operational efficiency. Ensured data quality and consistency across all projects using Python, Pandas, and Apache Airflow, reducing error rates by 20%. Automated testing processes with JUnit and Python, improving QA efficiency by 30% and ensuring solution reliability. Data & Machine Learning Engineer Lighthouse Consulting | Oman | Dec 2015 Apr 2017 Developed batch and real-time data pipelines using Kafka, PySpark, and AWS Glue, improving order processing efficiency by 50%. Delivered cloud-based solutions on AWS Lambda, EC2, and Redshift, enhancing system scalability and reliability. Spearheaded data reconciliation projects using Databricks, achieving 20% faster data ingestion for global clients. Built automated data dashboards in Tableau and ThoughtSpot, enabling faster executive decision-making by 20%. AI & Robotics Analyst Tata Consultancy Services (TCS) | India | Jan 2010 Jun 2015 Co-invented a context-based conversational AI system, earning a patent (US20160283463A1) for innovative NLP and ML techniques. Optimized natural language processing workflows with TensorFlow and NLTK, improving model accuracy and response times. Led unit testing and performance tuning using JUnit, Mockito, and SonarQube, reducing bugs and enhancing code quality. Created reusable code libraries in C++ and Java, streamlining development and improving project efficiency. Key Achievements Demand Forecasting: Delivered 90%+ accuracy in predictive models, optimizing inventory management and reducing costs. Fraud Detection: Reduced fraudulent activities by 30% with real-time anomaly detection models. Deployment Efficiency: Improved deployment cycles by 25% through streamlined CI/CD processes. Scalable Data Processing: Enhanced ETL processing speeds by 30% using Spark and Kafka pipelines. Patent: Co-invented and received a patent for a context-based conversational system (US20160283463A1). Education Master s in Management Studies (MMS) Finance Specialization Mumbai University, India | 2012 2014 Bachelor of Engineering (BE) Electronics & Telecommunications Mumbai University, India | 2005 2009 Certifications AWS Certified Machine Learning Specialty Certified Big Data Engineer (Cloudera/Hortonworks equivalent) Tableau Desktop Specialist Keywords: cplusplus continuous integration continuous deployment quality analyst artificial intelligence machine learning business intelligence sthree Alabama Colorado