AI/ML Engineer

Sai Laxman

**********@*****.*** +1-469-***-****

Senior AI/ML Consultant

PROFESSIONAL SUMMARY

•AI/ML professional with 8+ years of overall IT experience, transitioning from a strong software engineering foundation into building, deploying, and operationalizing machine learning and generative AI solutions for enterprise clients across telecom, financial services, healthcare, and media industries.

•Deep, hands-on command of the full ML lifecycle — from data collection, cleaning, and feature engineering through model training, validation, deployment, and monitoring — using Python, Pandas, NumPy, Scikit-learn, TensorFlow, and PyTorch.

•Proven experience designing and shipping production-grade Large Language Model (LLM) applications, including retrieval-augmented generation (RAG) pipelines, agentic workflows, prompt engineering, and fine-tuning (LoRA/QLoRA) using OpenAI, Azure OpenAI, Anthropic Claude, and Hugging Face models.

•Skilled in implementing/building vector database solutions (Pinecone, FAISS, Chroma, Weaviate) to power semantic search, enterprise chatbots, and knowledge-assistant platforms for business and support teams.

•Extensive experience across cloud-native ML environments, including AWS SageMaker, Bedrock, Azure Machine Learning, Lambda, S3, and container orchestration with Docker and Kubernetes.

•Strong track record driving cross-functional collaboration with data engineers, product owners, and business stakeholders to convert ambiguous problems into scalable, production-ready ML solutions within Agile/Scrum teams.

•Solid engineering background building data pipelines and backend services (Node.js, REST/GraphQL APIs, SQL/NoSQL databases) that integrate ML models into production applications end-to-end.

•Master’s degree in data science combined with real-world leadership in model development, MLOps/LLMOps practices, and cross-functional delivery, with a consistent focus on clean, maintainable code and reliable, well-tested pipelines.

TECHNICAL SKILLS

Programming Languages

Python, SQL, JavaScript, TypeScript, R

ML / DL Frameworks

Scikit-learn, TensorFlow, Keras, PyTorch, XGBoost, LightGBM

Generative AI / LLM

OpenAI GPT-4/GPT-4o, Azure OpenAI Service, Anthropic Claude API, LangChain, LangGraph, LlamaIndex, Hugging Face Transformers, Prompt Engineering, Retrieval-Augmented Generation (RAG), Fine-Tuning (LoRA/QLoRA/PEFT), Agentic AI Workflows (CrewAI, AutoGen), Model Context Protocol (MCP)

Vector Databases & Search

Pinecone, FAISS, Chroma, Weaviate, Qdrant, Semantic Search

MLOps & LLMOps

MLflow, Docker, Kubernetes, AWS SageMaker, vLLM, Triton Inference Server, Model Monitoring & Observability, LangSmith, RAGAS (LLM/RAG Evaluation), CI/CD Pipelines (GitHub Actions, Jenkins)

Cloud Platforms

AWS (S3, EC2, Lambda, SageMaker, Bedrock), Microsoft Azure (Azure ML, Azure OpenAI Service, Azure AI Studio), Google Cloud Platform (Vertex AI)

Data Engineering

Pandas, NumPy, Apache Spark, Apache Kafka, Apache Airflow, Databricks, ETL Pipeline Design, Feature Engineering, Feature Stores (Feast)

Databases

MySQL, PostgreSQL, MongoDB, Snowflake

Visualization & BI

Power BI, Tableau, Matplotlib, Seaborn

Backend & APIs

Node.js, Express.js, RESTful APIs, GraphQL

Methodologies & Tools

Agile, Scrum, Jira, Confluence, Git, GitHub, Terraform, Responsible AI Practices, Code Reviews, Sprint Planning

PROFESSIONAL EXPERIENCE

Comcast, Texas Nov 2025 – Present

Senior AI/ML Consultant

•Design and deliver machine learning components for the customer self-service portal, driving personalization, usage prediction, and support-automation initiatives in close partnership with data science and platform engineering teams.

•Design and maintain resilient data pipelines that ingest customer usage, billing, and account data into feature stores powering downstream ML models.

•Drive development of a generative AI assistant that enables customer service agents to instantly surface billing and account policy information, leveraging retrieval-augmented generation (RAG) over internal knowledge bases.

•Build and orchestrate agentic workflows with LangChain and LangGraph on Azure OpenAI to prototype conversational flows, fine-tune prompts, and rigorously evaluate response quality prior to handoff to the broader AI platform team.

•Implement vector search integration using Pinecone within the knowledge assistant, ensuring responses stay grounded in current product and policy documentation rather than relying purely on the model's own knowledge.

•Partner with data engineers to clean, validate, and prepare structured and semi-structured account data that drives churn and usage-pattern models.

•Manage end-to-end model deployment on AWS SageMaker, including packaging, endpoint configuration, and monitoring/alerting frameworks with the platform team.

•Develop Python scripts and notebooks for data exploration, feature engineering, and rapid model iteration, refactoring high-value components into production-grade code with the ML engineering team.

•Translate ambiguous business requirements — such as identifying customers who need proactive support — into actionable ML solutions in partnership with product owners, UX, and backend engineers.

•Contribute to code review standards across applications and ML notebook code, strengthening consistency of shared utilities and pipelines.

•Leverage GitHub Copilot and generative AI tools to accelerate data-processing development, debug pipeline issues, and resolve complex model-behavior edge cases.

•Strengthen CI/CD for ML services using GitHub Actions and Jenkins, streamlining reliable pipeline and model releases.

•Author comprehensive documentation of data sources, model assumptions, and known limitations, enabling cross-team collaboration and knowledge continuity.

Environment: Python, TensorFlow, PyTorch, Scikit-learn, LangChain, LangGraph, Azure OpenAI, Pinecone, AWS SageMaker, Docker, GitHub Actions, Jenkins, SQL, Node.js, REST APIs, GraphQL, Agile, Jira, GitHub

Consumers Credit Union, Chicago, IL Jul 2024 – Nov 2025

Senior AI/ML Consultant

•Contributed to a fraud-pattern detection initiative, building and evaluating classification models on historical transaction data to flag suspicious activity for review.

•Built robust data preparation and feature engineering pipelines in Python (Pandas, NumPy), transforming raw transaction and account data into model-ready datasets.

•Designed and benchmarked Scikit-learn and XGBoost models for transaction risk scoring, systematically comparing performance against the existing rules-based approach.

•Collaborated with the data engineering team to source and validate data from core banking and account systems, resolving data quality issues at the source.

•Stood up an internal chatbot proof-of-concept using Azure OpenAI to answer member-facing FAQ questions, driving prompt design and safety guardrails.

•Delivered integration between ML/AI prototypes and existing web applications built with React and Node.js, giving business teams direct visibility into early results.

•Partnered with backend and DevOps teams to containerize model-serving code with Docker and operationalize it in a shared testing environment.

•Participated in model evaluation discussions with risk and compliance stakeholders, translating technical metrics into clear, actionable explanations of model behavior and limitations.

•Migrated legacy reporting logic into Python-based analysis scripts, accelerating the analytics team's ability to iterate on new fraud rules.

•Actively contributed to Agile/Scrum ceremonies, including sprint planning and backlog grooming, owning estimates for data science and ML-related tasks.

•Tracked experiments, model versions, and results using MLflow, enabling rapid comparison across approaches and fast rollback when needed.

Environment: Python, Scikit-learn, XGBoost, Pandas, NumPy, Azure OpenAI, MLflow, Docker, AWS, React, Node.js, REST APIs, SQL, Agile, Jira, GitHub

Cigna Health, Indianapolis, IN Nov 2022 – Jun 2024

AI/ML Engineer

•Contributed to a predictive analytics project to identify members most likely to benefit from early outreach programs, working across claims and enrollment data.

•Built and cleaned datasets from multiple internal sources, resolving missing data, outliers, and inconsistent formats before feeding them into modeling pipelines.

•Developed and validated classification and regression models using Scikit-learn and TensorFlow/Keras to power risk-stratification and outreach-prioritization use cases.

•Evaluated model performance using precision, recall, and ROC-AUC, iterating on feature design to strengthen predictive accuracy.

•Built a location-search and member-matching feature that combined structured data lookups with a lightweight NLP-based matching approach.

•Delivered a React front end backed by Redux to surface model outputs — including risk scores and flags — to internal case managers in an actionable format.

•Wrote Python scripts for data extraction and transformation from REST APIs and internal databases, supporting both model training and reporting needs.

•Built TypeScript and Node.js services that exposed model predictions to internal applications, bridging the ML and application layers.

•Conducted early experiments applying pretrained NLP models from Hugging Face to classify and summarize unstructured case notes.

•Partnered with UI/UX, product owners, and clinical/business stakeholders to ensure model outputs were presented in a way that was genuinely useful for day-to-day workflows.

•Owned unit and integration testing across application and data pipeline code using Jest and Python testing tools, with peer-reviewed GitHub workflows.

Environment: Python, TensorFlow, Keras, Scikit-learn, Hugging Face Transformers, Pandas, NumPy, React, Node.js, TypeScript, REST APIs, SQL, MongoDB, Jest, GitHub, AWS, Agile, Jira

Warner Bros Jan 2022 – Oct 2022

AI/ML Engineer

•Contributed to a content-recommendation initiative for OTT streaming applications, leveraging viewing history and engagement data to surface relevant content to users.

•Built data pipelines to aggregate viewing, click, and session data from multiple platforms into a format usable for recommendation modeling.

•Designed collaborative-filtering and content-based recommendation approaches using Python and Scikit-learn alongside the data science team.

•Analyzed engagement and drop-off patterns to identify where content recommendations or UI changes could strengthen viewer retention.

•Integrated recommendation outputs into the React/Next.js front end, delivering personalized rows and suggestions to viewers.

•Aligned with backend and DevOps teams on API contracts for serving recommendation results and coordinated deployment timelines.

•Leveraged analytics and tracking tools to gather engagement data feeding both reporting dashboards and recommendation experiments.

•Cleaned and prepared historical viewing data, resolving duplicate records, inconsistent IDs, and missing metadata.

•Authored unit tests with Jest for front-end recommendation components, along with Python tests for data-processing scripts.

•Documented data definitions and modeling assumptions, enabling the wider analytics team to build on the work efficiently.

Environment: Python, Scikit-learn, Pandas, React, Next.js, TypeScript, REST APIs, SQL, Jest, GitHub, Jenkins, AWS, Agile, Jira, Video Streaming Platforms

Tech Mahindra, Hyderabad, India May 2018 – Jul 2021

Jr Data Scientist

•Delivered data-driven features for internal business applications, working with structured data from MySQL to build predictive and reporting tools.

•Developed Python scripts using Pandas and NumPy to clean, transform, and organize data from multiple internal systems for analysis and reporting.

•Built statistical models, rule-based scoring logic, and classification/regression models using Scikit-learn for business use cases such as demand estimation and customer segmentation.

•Conducted exploratory data analysis and created visualizations using Matplotlib and Seaborn to help stakeholders understand data trends and model results.

•Integrated data from REST APIs, third-party systems, and payment gateways into applications and modeling pipelines, ensuring accuracy and availability for analysis.

•Delivered machine learning outputs into web applications using React.js and Next.js, making predictions and scores accessible to business users.

•Handled MySQL query optimization, database design, and efficient data extraction to strengthen reporting and analytics performance.

•Partnered with business and UI/UX teams to translate requirements into clear, user-friendly presentations of data insights, predictions, and scoring results.

•Actively contributed to code reviews, Agile/Scrum ceremonies, testing, deployment, and production support across application, data, and model-related work.

•Built a strong foundation in SQL, Python, data analysis, and statistical modeling that progressed into formal machine learning engineering responsibilities in later roles.

Environment: PHP, Python, Pandas, NumPy, Matplotlib, React.js, Next.js, JavaScript, MySQL, HTML5, CSS3, REST APIs, GitHub, Jenkins, Agile

EDUCATION

Master of Data Science — University of North Texas, Texas

Bachelor of Computer Science — Malla Reddy Engineering College

Contact this candidate