Home

Prem Kumar - Gen AI / ML Engineer

[email protected]

Location: Horn Lake, Mississippi, USA

Relocation: Yes

Visa: OPT

PREM KUMAR Generative AI /ML Engineer Email: [email protected] PROFESSIONAL SUMMARY: Generative AI Engineer with over 6 years of experience in data science, machine learning, and AI architecture, specializing in natural language processing (NLP), large language models (LLMs), and cloud-based solutions. Proven expertise in developing, fine-tuning, and deploying AI models across diverse industries such as healthcare, banking, and retail. Proficient in Python, R, SQL, and advanced ML frameworks, with a strong focus on MLOps, prompt engineering, and innovative AI technologies to deliver scalable and impactful solutions. Automated model lifecycle management with advanced MLOps practices, including monitoring, versioning, and retraining. Developed end-to-end pipelines for data ingestion, preprocessing, and model deployment, ensuring reproducibility and efficiency. Utilized Python, R, and SQL for data wrangling, visualization, and analysis of structured and unstructured datasets. Collaborated with business stakeholders to align AI solutions with strategic objectives and measurable outcomes. Deployed scalable AI/ML solutions on cloud platforms like Azure OpenAI, AWS SageMaker, and Google AI, ensuring optimal performance. As a Generative AI/ML Engineer for a leading global provider of cloud healthcare solutions, I was responsible for designing and implementing AI-driven solutions to optimize healthcare processes. Designed and integrated generative AI applications using frameworks like TensorFlow, PyTorch, and Hugging Face. Built and fine-tuned AI chatbots, leveraging advanced prompt engineering to enhance conversational accuracy and relevance. Applied advanced machine learning and statistical techniques to extract actionable insights and improve decision-making. Worked in agile and scrum environments, optimizing workflows and fostering collaboration with onshore and offshore teams. Ensured robust CI/CD pipelines and seamless AI model deployment with tools like Azure DevOps, Jenkins, and GitHub Actions. Delivered deep learning solutions, including CNNs, RNNs, LSTMs, and transformers, for applications like time-series forecasting and image processing. Optimized model training and deployment on BigQuery, Vertex AI, and other cloud platforms using advanced methodologies. Engineered feature pipelines for structured and unstructured datasets, incorporating real-time and batch processing. Designed recommendation systems and personalization engines using collaborative filtering and deep learning techniques. Conducted A/B testing and model validation to ensure AI solutions meet performance and business criteria. Spearheaded initiatives to integrate explainable AI (XAI) methods, improving model transparency and trust. Managed data governance and compliance for AI workflows, ensuring adherence to regulatory standards. Leveraged transfer learning and fine-tuning for specialized tasks, reducing training time while maintaining accuracy. Conducted workshops and training sessions to upskill teams in AI tools, MLOps, and generative AI technologies. Championed cross-functional collaboration to streamline AI model development and deployment in cloud environments. EDUCATION DETAILS: Masters in computer science from Christian Brothers University, USA. TECHNICAL SKILLS: Programming Languages R, Python 2.X & 3.X (NumPy, SciPy, pandas, seaborn, beautiful soup, scikit-learn, NLTK), SQL, C Databases Oracle, , MySQL, SQL Server, MongoDB, Dynamo DB, Cassandra. Analytic Tools R 2.15 / 3.0 (Reshape, ggplot2, Dlpr, Car, Mass and Lme4), Excel, Data Studio Big-Data Framework Hadoop Ecosystem 2.X (HDFS, MapReduce, Hive 0.11, HBase 0.9), Spark Framework 2.X(Scala 2.X, SparkSQL, Pyspark, SparkR, Mllib) Version Control SVN, GIT, GitHub, Git 2.X CI/CD Tools Azure DevOps, ADO Operating Systems Windows 10/7/XP/2000/NT/98/95, UNIX, LINUX, OS Machine Learning Algorithms Linear regression, SVM, KNN, Naive Bayes, Logistic Regression, Random Forest, Boosting, K-means clustering, Hierarchical clustering, Collaborative Filtering, Neural Networks, NLP Data Visualization Tools Tableau 8.0 /9.2 / 10.0, Plotly, R-ggplot2, Python-Matplotlib, Logi Analytics WORK HISTORY: Client: Blue Health Intelligence, NJ Jan 2024 to till now Role Generative AI/ML Engineer Responsibilities: Developed in-depth expertise in Generative AI mathematics and practical applications, such as Retrieval-Augmented Generation (RAG) and Parameter Efficient Fine-Tuning (PEFT). Fine-tuned machine learning models using Hugging Face Transformers and PyTorch to address specific business challenges effectively. Designed, built, and deployed advanced AI models with TensorFlow, PyTorch, and Keras, solving complex operational and customer-centric problems. Created intelligent AI agents to streamline tasks and enhance overall business efficiency. Utilized vector databases like Pinecone and FAISS to optimize data retrieval and boost AI-driven insights. Managed large language models (LLMs) and fine-tuning pipelines through model registries such as Hugging Face. Built NLP models for use cases like sentiment analysis, text classification, and named entity recognition (NER) to derive actionable insights. Developed custom embeddings, tokenizers, and word vectors using spaCy and NLTK for robust NLP pipelines. Implemented AI-driven search and recommendation systems by integrating vector embeddings with advanced databases. Established end-to-end MLOps pipelines to ensure efficient deployment, monitoring, and maintenance of AI models. SQL server migration to Azure cloud databases for data warehousing for the goal of using Azure Machine Learning and MS R server predictive analytics in the cloud. knowledge in Big Data & Hadoop technologies involving HDFS, Sqoop, Hue, Hive, Impala, Spark (SQL), Pig, Hbase, NoSQL database, Kafka, Solr, Hbase, Flume, Oozie, StreamSets, NiFi, Datameer, Trifacta, Dataiku, Python and Linux Refined prompt engineering strategies to enhance the contextual accuracy and relevance of LLM outputs. Deployed scalable AI chatbots in cloud environments, improving operational performance and interaction quality. Wrote modular and efficient Python code for building machine learning algorithms and pipelines. Applied machine learning techniques such as Random Forest, LightGBM, and XGBoost for predictive modeling and classification. Worked on KAGGLE data sets and Microsoft Azure ML predictive models as a part of Data science Boot camp. Designed and validated diverse machine learning models, including regression, classification, and deep learning architectures, to address various business requirements. Leveraged advanced ML frameworks to ensure model scalability and optimal performance. Translated business objectives into analytical tasks, driving the development of AI/ML solutions aligned with organizational goals. Built predictive analytics solutions using MapReduce and Spark on AWS for scalable data processing. Identified and implemented innovative Generative AI solutions tailored to specific business needs and use cases. Analyzed large datasets to uncover patterns and actionable insights, leveraging strong analytical and problem-solving skills. Used Spark and PySpark for real-time analysis of financial data, such as loan default predictions. Created impactful data visualizations with Tableau to facilitate data-driven decision-making for stakeholders. Demonstrated expertise in Python and R for analytics, data mining, and model development, while integrating cloud-based platforms like AWS SageMaker and Snowflake. Kept up to date with advancements in Generative AI, data quality management, and AI/ML best practices to maintain a competitive edge. Client: John Deere Feb 2023 to Dec 2023 Role: AI/ML Engineer Responsibilities: Gained comprehensive expertise in the mathematics and practical applications of Generative AI, including Retrieval-Augmented Generation (RAG) and Parameter Efficient Fine-Tuning (PEFT) techniques. Performed supervised fine-tuning of ML models with Hugging Face Transformers and PyTorch to address tailored business needs. Designed, implemented, and deployed advanced AI models using TensorFlow, PyTorch, and Keras to solve complex operational and customer-facing challenges. Visualized data using different visualization tools R, Azure ML and Power BI Developed intelligent AI agents to automate tasks and optimize business processes for enhanced efficiency. Leveraged vector databases such as Pinecone and FAISS to optimize data retrieval and improve AI-driven insights. Managed large language models (LLMs) and fine-tuning workflows using model registries like Hugging Face. Built NLP models for sentiment analysis, text classification, and named entity recognition (NER) using cutting-edge techniques. Utilized latest technologies and rich ecosystem of tools provided by Hadoop such as HBase, Dataiku(Machine Learning), Hive, Kafka, Solr Worked on Azure databases, the database server is hosted on Azure and use Microsoft credentials to login to the DB rather than the Windows authentication that is typically used. Engineered custom tokenizers, embeddings, and word vectors with tools like spaCy and NLTK to create robust NLP solutions. Designed and deployed AI-powered search and recommendation systems using vector embeddings integrated with advanced databases. Implemented end-to-end MLOps pipelines for seamless AI model deployment, monitoring, and maintenance. Optimized prompt engineering strategies to improve the accuracy and contextual relevance of insights derived from LLMs. Developed scalable cloud-based AI chatbots, improving response quality and operational efficiency. Wrote modular, high-performance Python code to implement ML algorithms and data pipelines effectively. Applied machine learning algorithms, including Random Forest, LightGBM, and XGBoost, for predictive analytics and classification tasks. Created and validated diverse ML models, including regression, classification, and deep learning architectures, to solve complex business problems. Used advanced ML frameworks to improve model scalability and ensure high performance. Translated business requirements into analytical objectives, leading the design of AI/ML solutions aligned with organizational goals. Developed predictive analytics modules with MapReduce and Spark on AWS infrastructure for scalable insights. Proposed and implemented innovative Generative AI solutions tailored to business needs, identifying impactful use cases. Used librosa, scipy.io to analyze audio signals and to generate two individual mono wav files from a two-channel stereo file. Designed, consulted - on, and implemented Azure data analytics/data science/data warehousing platform solution architectures solving capabilities. Used Spark and PySpark to perform real-time analysis of critical financial metrics, such as loan defaults. Designed data visualizations with tools like Tableau, enabling stakeholders to make informed, data-driven decisions. Showcased proficiency in Python and R for data mining, analytics, and model development, integrating cloud platforms like AWS Sage Maker and Snowflake. Stayed up to date with emerging trends in Generative AI, data quality management, and AI/ML practices to remain ahead of industry advancements. Analyzed large datasets to identify patterns and derive actionable insights, demonstrating strong problem- Client: Innominds Software Pvt Ltd, Hyderabad, India. Sep 2018 to Dec 2021 Role: Data Scientist Responsibilities: Designed and executed strategies for seamless integration between legacy systems and Snowflake/SAS Viya, enabling enhanced data processing and operational efficiency. Ensured adherence to federal, state, and organizational data governance regulations by implementing stringent data security and privacy protocols. Developed and automated ETL pipelines using AWS Glue and Lambda, resulting in significantly improved data workflow efficiency. Collaborated with stakeholders, including regulatory bodies, to define compliance reporting requirements and ensure accurate and timely data submissions. Performed comprehensive data audits and validations to maintain high data quality standards during migration processes. Incorporated robust error-checking and validation mechanisms within ETL workflows to proactively detect and resolve discrepancies. Explored data and plotted features using Python packages like LibROSA and Matplotlib. Optimized data transformations, queries, and reporting in Snowflake and SAS Viya, improving processing speed and system performance. Managed data assets using tools like AWS Code Commit, S3 Buckets, DynamoDB, and Glue Crawlers to enable efficient data cataloging and accessibility. Participated in Agile ceremonies, ensuring timely completion of data migration tasks and reporting functionalities. Diagnosed and resolved data integrity issues through root cause analysis using ETL processes, SQL, Python, and R. Implemented proactive solutions to address recurring data quality issues, reducing future risks and enhancing system reliability. Client: GSS INFOTECH, Hyderabad, India. Nov 2017 to Aug 2018 Role: Data Engineer Responsibilities: Mainly engaged in data migration processes utilizing Cloudera, integrated with Bitbucket repository and CI/CD Replicated existing application logic and functionalities within Azure Data Lake, Data Factory, SQL Database, and SQL Data Warehouse environments Proficient in Azure Cloud Services spanning PaaS & IaaS, including Azure Synapse Analytics, SQL, Azure Data Factory, Azure Analysis Services, Application Insights, Azure Monitoring, Key Vault, and Azure Data Lake Managed Git repositories on Bitbucket, enforcing best practices for branch management, code review, and merge strategies to maintain code quality and project integrity Conducted workload migrations from on-premises systems to Microsoft Azure leveraging Azure Site Recovery Developed R-based data analysis and visualization solutions using RStudio, leveraging its integrated development environment (IDE) and rich ecosystem of packages and libraries Contributed to the design of data warehouses using Star-Schema methodology and converted data from diverse sources into SQL tables Implemented Flyway for version-controlled database migration ensuring consistency and reliability Integrated Flyway with CI/CD pipelines to automate the deployment of database changes Applied various machine learning algorithms and statistical modeling techniques such as decision trees, regression models, neural networks, SVM, and clustering using the Scikit-learn package in Python Utilized Spark to structure large volumes of unstructured data from multiple sources. Integrated Docker with CI/CD pipelines to automate the build, test and deployment processes Collaborated within a team on a large-scale project, employing SQL to interface with databases to fetch and comprehend data storage and management. Extensive hands-on development experience in Python and ML model development Designed, developed, and implemented data discovery dashboards in Tableau to effectively present data insights to business stakeholders Evaluated additional data inputs and methodologies to enhance model results and identified opportunities for improvement Keywords: cprogramm continuous integration continuous deployment artificial intelligence machine learning business intelligence sthree database rlang microsoft New Jersey