Machine Learning Engineer

Om Sai Kiran

Sr. Machine Learning Engineer

************@*****.***

***- ***-****

https://www.linkedin.com/in/omsaikiran-a-38a618145/

Professional Summary:

Over 10+ years of IT industry experience encompassing Machine Learning, Data Mining with large datasets of structured and unstructured data, Data Acquisition, Data Validation, Predictive Modeling, and Data Visualization.

Extensive experience in Text Analytics, developing statistical machine learning and data mining solutions to various business problems, and generating data visualizations using R, Python, and Tableau.

Over 5+ years of experience with machine learning techniques and algorithms such as k-NN, Naive Bayes, Random Forest, SVM, XGBoost, Logistic Regression, Linear Regression, Lasso Regression, K-Means Clustering, and Neural Networks.

Hands-on experience in foundational and advanced machine learning models including regression, boosting (GBM, XGBoost), decision trees, Random Forests, Neural Networks (NNs), Hidden Markov Models (HMMs), Conditional Random Fields (CRFs), and Markov Random Fields (MRFs).

Highly skilled in using data visualization tools such as Tableau, ggplot2, d3.js, QlikView, and Anaplan for dashboards and forecasting.

Strong proficiency in statistical and programming tools/languages such as R, Python, C, C++, Java, SQL, UNIX, MATLAB, Scala, Perl, and SPSS.

Experienced in MLOps practices, implementing Docker, Kubernetes, AWS SageMaker, GitLab CI/CD pipelines, and GitHub Actions to automate and streamline ML workflows, model retraining, and deployment.

Specialized in Large Language Models (LLMs) like GPT, BERT, T5, Llama2, and Mistral with hands-on experience fine-tuning pre-trained models to adapt them for specific use cases and improve real-world application performance.

Proficient in designing and deploying AI chatbots and intelligent agents using frameworks such as LangChain and LangGraph, and implementing Agentic AI architectures for dynamic decision-making and multi-step task execution.

Built and deployed LSTM neural networks for text data (item descriptions, comments), deep neural networks integrating LSTM outputs with additional features, and applied deep learning for image captioning and stock prediction.

Familiar with CNN, RNN, autoencoders, and hybrid CNN-RNN architectures for deep learning applications in image and sequence processing.

Implemented AI and deep learning platforms/methodologies using TensorFlow, PyTorch, RNN, LSTM, and data flow graphs.

Experienced in extracting and engineering Value Added Datasets using Python, R, Azure, and SQL for targeted customer segmentation, behavioral analysis, and actionable insights.

Built recommendation systems using collaborative filtering, autoencoders, and clustering; applied dimensionality reduction techniques like PCA.

Hands-on experience with NoSQL databases including HBase, Cassandra, and MongoDB, as well as traditional RDBMS (Oracle, MS SQL Server, DB2, Teradata).

Proficient in integrating and transforming data from relational databases, flat files, and various sources into staging, ODS, Data Warehouses, and Data Marts.

Experience in software development life cycle (SDLC) with strong familiarity in both Waterfall and Agile/Scrum methodologies.

Experienced in data modeling and design using tools like Erwin, Power Designer, and ERStudio.

Skilled in creating analytics, OLAP reports, PivotTables, Ad-hoc reporting, graphs, and dashboards.

Adept at using Boto3 for managing AWS resources such as S3 (for embedding storage) and Lambda (for model inference or task execution), EC2.

Familiar with developing intelligent AI-driven applications in smart factory environments, leveraging AI and ML for real-time analytics and hyper automation.

Strong foundation in statistics, mathematics, recommendation algorithms, business analytics, and effective use of data science for operational insights.

Technical skills:

Programming Languages: Python, R, C.

Frameworks & Tools: TensorFlow, Porch, LangChain, Langgraph, Llama Index, Stream lit, Flask, Django, Scikit-learn, Hugging Face Transformers, OpenCV, Apache Airflow, Apache Spark, Hadoop (HDFS, Hive, MapReduce, Pig), Flow, Pandas, NumPy, Matplotlib, Seaborn.

Generative AI Technologies: Llama2, Mistral, OpenAI, Google Gemini Pro (both open-source and paid LLM models), Large Language Models (LLMs) like GPT, BERT, and T5, Generative Adversarial Networks (GANs), VAE’s, Retrieval-Augmented Generation (RAG)

Natural Language Processing (NLP): Hugging Face, SpaCy, NLTK, Transformers, Named Entity Recognition (NER), Relation Extraction, Text Generation, Summarization, Translation, Prompt Engineering

Vector Databases: Chroma DB, Pinecone, FAISS

Database Management: DataStax Cassandra DB (production environments), MySQL, MongoDB, SQL Server, PostgreSQL, RDS, Redshift

Deployment Platforms: AWS Bedrock, AWS (EC2, Lambda, S3, Sage Maker), Azure Functions, Hugging Face Spaces, Google Cloud, Microsoft Azure AI, Docker Hub, Kubernetes

AI/ML Techniques: Fine-tuning with custom data, vector embedding, NLP, neural network optimization, MLOps, Hyperparameter Tuning, A/B Testing, Ensemble Learning, Transfer Learning, Recommender Systems, Gradient Boosting Machines, Random Forest, LSTM, Recurrent Neural Networks, Char CNNs, Logistic Regression, SVM, K-means Clustering, Hierarchical Clustering, Collaborative Filtering, Predictive Modeling, Regression Analysis

Data Science: Exploratory Data Analysis (EDA), Statistical Modeling, Feature Engineering, Data Preprocessing, Data Pipelines, Time Series Analysis

Data Integration & APIs: RESTful APIs, GraphQL, WebSocket’s, Real-time Data Pipelines, ETL processes

Data Governance: Data Privacy, Security, Compliance, Ethics in AI, Fairness in AI, Model Explainability

Version Control: Git, GitHub

Professional Experience:

Advance Auto Parts May 2022 – Present date

Role: Sr. Machine Learning Engineer

Responsibilities:

Led architecture and deployment of GenAI solutions using AWS (SageMaker, Lambda, Bedrock), Azure ML Studio, and GCP Vertex AI, ensuring scalability, observability, and CI/CD integration

Provided architectural leadership in designing and scaling AI-first enterprise solutions, with a strong focus on integrating LLMs, traditional ML models, and cloud-native AI pipelines.

Architected and deployed end-to-end Generative AI solutions by integrating LLMs into enterprise workflows, enabling intelligent automation and autonomous decision-making at scale.

Led the design and deployment of large language model frameworks including GPT-3.5 Turbo, Hugging Face Transformers, and LLaMA-based models, powering document summarization, chat interfaces, and decision-support systems.

Automated infrastructure provisioning using Terraform modules to deploy AWS services (S3, SageMaker, Lambda, EKS) in a reproducible and secure manner.

Designed and maintained Terraform-based blueprints for scalable, cost-optimized ML environments following infrastructure-as-code best practices.

Developed and maintained CI/CD pipelines for ML models using GitHub Actions, AWS Code Pipeline, and Docker, enabling consistent and auditable model deployment.

Integrated model testing, validation, containerization, and version control into automated CI workflows to accelerate model release cycles

Implemented Agentic AI frameworks using Lang Graph and AutoGen to build multi-agent systems capable of dynamic task orchestration, adaptive reasoning, and self-directed goal execution.

Fine-tuned BERT models for NLU tasks, including named entity recognition (NER) and intent classification, improving accuracy and context awareness in voice and chatbot-based applications.

Applied GPT-3.5 Turbo with advanced prompt engineering to enable zero-shot and few-shot learning for use cases such as Text-to-SQL, document analysis, and enterprise knowledge extraction.

Engineered retrieval-augmented generation (RAG) pipelines using LangChain, Chroma DB, and Llama Index, combining semantic search and contextual generation to optimize question-answering and content retrieval.

Implemented Agentic AI frameworks using Lang Graph and AutoGen, orchestrating multi-agent collaboration for healthcare triage, claims processing, and autonomous task planning.

Conducted rigorous A/B testing and hyperparameter tuning to benchmark and improve GenAI model outcomes for internet-scale applications.

Delivered Generative AI capabilities (chatbots, summarization tools, Q&A systems) for high-traffic web applications, improving user engagement and self-service resolution rates by over 40%.

Developed robust chatbot frameworks with GPT-3.5 Turbo and LangChain, enabling multi-turn dialogue, session memory, and real-time decision support integrated with enterprise APIs.

Applied advanced prompt engineering strategies for structured output formatting, data extraction, and conversation grounding, enhancing the interpretability and control of LLM responses.

Integrated traditional machine learning models such as Random Forest, Support Vector Machines (SVM), and Logistic Regression to support fraud detection, churn prediction, and classification tasks within hybrid AI systems.

Built and deployed scalable deep learning models using PyTorch, TensorFlow, and Scikit-learn, optimizing both LLMs and classical models in ensemble architectures.

Delivered complete MLOps pipelines to support training, deployment, and monitoring using Docker, Kubernetes, SageMaker, and GitHub Actions, reducing model onboarding time by 30%.

Designed and orchestrated end-to-end ML workflows using AWS SageMaker Pipelines, automating data ingestion, preprocessing, model training, evaluation, and deployment; enabled CI/CD for ML models with integrated model registry and approval workflows, improving deployment consistency and reducing manual overhead by 40%.

Utilized AWS Bedrock and SageMaker for scalable deployment of foundation models, automating use cases like claims summarization, document classification, and contextual Q&A.

Created NLU pipelines to extract structured meaning from unstructured text using BERT, CRFs, and rule-based systems, enhancing downstream chatbot and analytics workflows.

Built and orchestrated ML workflows using Apache Airflow and SageMaker Pipelines, streamlining feature engineering, model training, and deployment with reusable templates.

Evaluated and integrated Kubeflow Pipelines for managing large-scale ML workflows across teams, improving cross-functional experimentation and traceability.

Developed real-time data pipelines using Google Pub/Sub, Dataflow, and TensorFlow Serving, supporting low-latency inference for recommendation engines and customer engagement models.

Integrated Big Query ML and Looker Studio with GenAI models to provide interactive dashboards, sentiment analysis, and executive-level insights powered by LLM summarization.

Collaborated with cross-functional teams in Agile sprints to embed BERT, GPT, and traditional ML models into production services, enabling intelligent automation and customer interaction at scale.

Built customer segmentation models using K-means clustering, PCA, and hierarchical clustering, enhancing targeting and personalization strategies in marketing and digital engagement.

Created MapReduce/Spark-based Python modules for predictive modeling and distributed training, including custom implementations of Random Forests and regression models using Hadoop Streaming.

Designed hybrid AI systems that combine deep learning models and rule-based engines, ensuring compliance, traceability, and interpretability in regulated industries.

Automated ETL workflows and data preprocessing using Spark, AWS Glue, and shell scripting, enabling efficient data preparation for training LLMs and traditional models.

Leveraged by CLI scripting for bulk data ingestion into Big Query, streamlining preprocessing pipelines for downstream AI modeling and reporting.

Partnered with cross-functional teams including data engineers, software architects, and product managers to embed AI components into customer-facing digital products and backend systems.

Continuously conducted hyperparameter tuning and A/B testing, optimizing both LLM and traditional model performance while reducing inference latency by 25%.

Net Cracker Technologies Jan 2019 - March 2022

Role: Sr. Machine Learning Engineer

Responsibilities:

Developed and maintained AI/ML applications using Python and integrated with cloud-based AI services on AWS, GCP, and Azure to deliver scalable and efficient solutions.

Developed generative NLP applications using GPT-2 and GPT-3 APIs to enhance autonomous language understanding and decision-making workflows.

Utilized Apache Spark, Scala, Hadoop, HBase, Fire store, MongoDB, Pub/Sub, Dataflow, Spark Streaming, and MLlib, leveraging a wide variety of machine learning methods including classification, regression, and dimensional reduction to optimize user engagement and performance.

Built AI agent workflows using Knowledge Graphs and symbolic reasoning to enhance natural language understanding and applying early transformer models for contextual response generation.

Designed and deployed intelligent chatbot systems using LLMs (GPT-2/GPT-3), integrating contextual awareness, multi-turn conversation logic, and dynamic response generation.

Extensive experience in natural language processing (NLP) and computer vision (YOLOv4/v5 and OpenAI’s CLIP (2021)) applications.

Deployed containerized ML inference services using Amazon EKS, ensuring high availability and scalability of production models across multiple environments.

Created Helm charts and Kubernetes manifests for managing ML microservices and ensuring efficient autoscaling and monitoring.

Implemented MLOps pipelines using Vertex AI Pipelines, Kubeflow, and Cloud Build to ensure seamless model deployment, monitoring, and retraining processes.

Automated data ingestion, transformation, and feature extraction pipelines using Python, integrating with Big Data technologies like Apache Spark, Hadoop, and Kafka.

Developed and optimized multi-modal AI systems (text, image, audio, video) using state-of-the-art deep learning techniques and integrated real-time data through Pub/Sub and Dataflow.

Integrated cloud-based AI solutions with services on AWS, GCP, and performed model tuning for optimal performance and efficiency.

Championed AI system integration strategies to bridge LLMs, traditional ML, and business logic across complex data ecosystems.

Designed resilient AI pipelines with real-time data streaming, model versioning, automated retraining, and governance using Kubernetes, Docker, Moldflow, and SageMaker Pipelines.

Worked on data analysis and pipeline integration using Big Query, Cloud SQL, and Fire store, enhancing data accessibility and real-time processing capabilities.

Designed AI system architectures using AWS Bedrock and Azure ML Studio for scalable GenAI model deployment and real-time inference across cloud-native platforms.

Developed intelligent agents using LangChain that autonomously retrieve information, trigger workflows, and make context-aware decisions without human intervention.

Programmed utilities in Python for data preprocessing, data mining, and analysis, integrating with cloud-native services for scalable operations.

Worked with data modeling tools like Erwin Data Modeler and UML (Unified Modeling Language) using Visio to design conceptual, logical, and physical data models.

Built context-aware conversational AI experiences using GPT-3.5/4 and LangChain integrated with enterprise APIs for seamless interaction in SaaS platforms.

Implemented Agile methodology to manage project development, collaborating closely with business analysts, SMEs, and data architects to align technical solutions with business needs.

Designed and maintained OLAP Databases/Cubes, Scorecards, Dashboards, and Reports using Big Query and Looker Studio for comprehensive data visualization and reporting.

Developed end-to-end MLOps pipelines with Vertex AI, Kubeflow, and Cloud Build, ensuring continuous integration and automated model updates.

Utilized Data Quality Validation techniques to validate Critical Data Elements (CDE) and identify anomalies using Google Data Catalog and Data plex.

Enhanced model performance through compression and quantization using TensorFlow Lite, Pytorch and Vertex AI, reducing inference time by 30%.

Built connectors for integrating data from APIs, web scraping tools, and databases into LLM workflows using Python, Cloud Functions, and Pub/Sub.

Implemented classification and regression algorithms using Logistic Regression, Decision Trees, KNN, and Naive Bayes via Vertex AI and Scikit-learn.

Used F-Score, Precision, Recall, and A/B testing to evaluate Model performance.

Designed and optimized data storage and retrieval mechanisms using Google Cloud Storage, Big Query, and Dataflow to support large-scale machine learning workloads.

Developed Python -based ETL pipelines to load and transform data from various sources, including flat files, Cloud SQL, and Big Query.

Applied deep learning techniques for NLP tasks, including text classification, sentiment analysis, and entity recognition, leveraging LLMs for enhanced language understanding.

Conducted Gap Analysis to identify performance bottlenecks and opportunities for improvement in AI models and workflows.

Designed and developed UML diagrams including Use Case, Activity, Sequence, and Class Diagrams to visualize system architecture and workflows.

Conducted research on emerging AI technologies and frameworks to enhance the capabilities of generative and autonomous AI systems.

Demonstrated experience in cloud-based data processing and ETL pipelines, leveraging Python and cloud-native technologies to build robust data integration workflows.

Hinduja Global Solutions Jan 2016 – Jan 2019

Role: Sr. Machine Learning Engineer

Responsibilities:

Used Pandas, NumPy, Seaborn, SciPy, Matplotlib, Scikit-learn, and NLTK in Python for developing various machine learning algorithms.

Worked on different data formats such as JSON, XML and performed machine learning algorithms in Python.

Setup storage and data analysis tools in Amazon Web Services cloud computing infrastructure.

Built and deployed machine learning models using AWS SageMaker, leveraging built-in algorithms for classification and regression. Used SageMaker Pipelines to automate model training, validation, and deployment, reducing operational overhead by 30%.

Implemented end-to-end systems for Data Analytics, Data Automation and integrated with custom visualization tools using R, Mahout, Hadoop and MongoDB.

Utilized Spark, Scala, Hadoop, HBase, Cassandra, MongoDB, Kafka, Spark Streaming, MLLib, Python, a broad variety of machine learning methods including classifications, dimensionality reduction etc. and utilized the engine to increase user lifetime by 45% and triple user conversations for target categories.

Used Spark Data frames, Spark-SQL, Spark MLLib extensively and developing and designing POC's using Scala, Spark SQL and MLlib libraries.

Applied deep learning techniques using TensorFlow, Kera’s, and PyTorch to improve model performance and prediction accuracy.

Used Data Quality Validation techniques to validate Critical Data Elements (CDE) and identified various anomalies.

Participated in all phases of Data mining, Data-collection, Data-Cleaning, Developing-Models, Validation, Visualization and Performed Gap Analysis.

Data Manipulation and Aggregation from different sources using Nexus, Toad, Business Objects and SmartView.

Implemented Agile Methodology for building an internal application.

Good knowledge of Hadoop Architecture and various components such as HDFS, Job Tracker, Task Tracker, Name node, Data node, Secondary Name node, and MapReduce concepts.

Programmed a utility in Python that used multiple packages (SciPy, NumPy, Pandas) Implemented Classification using supervised algorithms like Logistic Regression, Decision trees, KNN, Naive Bayes.

Updated Python scripts to match data with our database stored in AWS Cloud Search, so that we would be able to assign each document a response label for further classification.

Created SQL tables with referential integrity and developed queries using SQL, SQL PLUS and PL/SQL.

Interaction with Business Analyst, SMEs and other Data Architects to understand Business needs and functionality for various project solutions

Identifying and executing process improvements, hands-on in various technologies such as Oracle and Business Objects.

KPMG Technological Services Aug 2014 – Jan 2016

Role: Data Scientist

Responsibilities

Excellent experience in implementing Machine Learning methods like Optimization and Visualization. Mathematical methods of statistics such as Regression Models, Decision Tree, Naïve Bayes, Ensemble Classier, Hierarchical Clustering and Semi-Supervised Learning on different datasets using Python.

Involved in migrating 1.6 trillion equity related Raw data over the period of 8 weeks to different Hadoop clusters and provide batch wise predictions using regression algorithms such as Ridge Regression, Lasso Regression and Weighted Averaging Ensemble techniques.

Hands on experience in stockpiling and information examination apparatuses in Amazon Web Services (AWS) distributed computing framework.

Built and deployed scalable ML models on AWS SageMaker, leveraging built-in algorithms for classification, regression, and anomaly detection. Used SageMaker Pipelines to automate model training, evaluation, and deployment, improving prediction efficiency by 25%.

Excellent working experience in building predictive models including Support Vector Machine, Decision tree, Naive Bayes Classier, Neural Network plus ensemble methods of the models to evaluate how the likelihood to recommend of customer groups would change in different sets of services by using python (Scikit-learn) & Jupiter notebook and pushed them in production using Python’s pickle package.

Worked with a few R bundles including knitr, dplyr, SparkR, Causal Infer, Space-Time, Interface with Caffe Deep Learning Framework.

Proficient in CI/CD pipelines and tools, including Jenkins, Spinnaker, Bitbucket, Splunk, CloudWatch, Grafana, Dynatrace, Terraform, etc.

Implementing investigation calculations in Python. Pandas, NumPy, Seaborn, SciPy, Matplotlib, Sci-pack Learn, and NLTK in Python for creating different AI calculations.

Excellent experience in planning and usage of Statistical models, Predictive models, undertaking information model, metadata arrangement and information life cycle the executives in both RDBMS, Big Data conditions.

Hands-on experience Natural Language models and Fortification learning motors to Streamline Clever Specialists that Mechanize Task Execution.

Developed and designed k-Means bunching to comprehend client ordered purchased items and portion the clients dependent on the client items for creature medication and immunizations conduct data for redid item offering, modified and need administration, to improve existing gainful connections and to stay away from client agitate, and so forth utilizing Python.

Applied Text investigation on unstructured email information utilizing Natural language preparing toolbox (NLTK).

Experience in extracting features from textual data using recurrent neural network techniques like LSTM, GRU and BERT.

Worked with content to Vector portrayal techniques including Counter Vectorizer, Tf-idf for point demonstrating.

Experienced in data scraping using PySpark Machine learning library to build and evaluate different models.

Involved in implementing data analysis with various analytic tools, such as Anaconda 4.0 Jupiter Notebook 4.X, R 3.0 (Caret, dplyr) and Excel.

Experienced in Agile Methodology creating various Proof of Concepts (PoC) and Gap analysis and gathered necessary data for analysis from different sources, prepared data for data exploration using data munging.

Worked on different data sets like JSON, XML and performed AI calculations in Python.

Contact this candidate