Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Hi, I’m Sri Premanth Pasumarthi, an AI/ML engineer with a Master’s in Computer Science and 5+ years of experience building scalable data pipelines, deploying ML models, and delivering real-time scoring systems. I enjoy turning complex data into actionable insights and collaborating with data science, IT, and business teams to drive measurable outcomes. I’m passionate about designing end-to-end solutions—from data ingestion and feature engineering to model monitoring and production-grade deployments on cloud-native platforms. I thrive in cross-functional roles, continuously learning new tools and techniques to improve model health, data readiness, and user impact.…Hi, I’m Sri Premanth Pasumarthi, an AI/ML engineer with a Master’s in Computer Science and 5+ years of experience building scalable data pipelines, deploying ML models, and delivering real-time scoring systems. I enjoy turning complex data into actionable insights and collaborating with data science, IT, and business teams to drive measurable outcomes. I’m passionate about designing end-to-end solutions—from data ingestion and feature engineering to model monitoring and production-grade deployments on cloud-native platforms. I thrive in cross-functional roles, continuously learning new tools and techniques to improve model health, data readiness, and user impact.

Sri Premanth Pasumarthi

AI Engineer, Data Scientist, Full Stack Developer, +2





Hi, I’m Sri Premanth Pasumarthi, an AI/ML engineer with a Master’s in Computer Science and 5+ years of experience building scalable data pipelines, deploying ML models, and delivering real-time scoring systems. I enjoy turning complex data into actionable insights and collaborating with data science, IT, and business teams to drive measurable outcomes. I’m passionate about designing end-to-end solutions—from data ingestion and feature engineering to model monitoring and production-grade deployments on cloud-native platforms. I thrive in cross-functional roles, continuously learning new tools and techniques to improve model health, data readiness, and user impact.…Hi, I’m Sri Premanth Pasumarthi, an AI/ML engineer with a Master’s in Computer Science and 5+ years of experience building scalable data pipelines, deploying ML models, and delivering real-time scoring systems. I enjoy turning complex data into actionable insights and collaborating with data science, IT, and business teams to drive measurable outcomes. I’m passionate about designing end-to-end solutions—from data ingestion and feature engineering to model monitoring and production-grade deployments on cloud-native platforms. I thrive in cross-functional roles, continuously learning new tools and techniques to improve model health, data readiness, and user impact.

Available to hire

Hi, I’m Sri Premanth Pasumarthi, an AI/ML engineer with a Master’s in Computer Science and 5+ years of experience building scalable data pipelines, deploying ML models, and delivering real-time scoring systems. I enjoy turning complex data into actionable insights and collaborating with data science, IT, and business teams to drive measurable outcomes.

I’m passionate about designing end-to-end solutions—from data ingestion and feature engineering to model monitoring and production-grade deployments on cloud-native platforms. I thrive in cross-functional roles, continuously learning new tools and techniques to improve model health, data readiness, and user impact.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Work Experience

AI/ML Engineer at TCS-USAA

November 1, 2024 - Present

Automated data ingestion and model execution workflows using Apache Airflow DAGs and Control-M, including upstream dependency management and downstream scoring job orchestration to meet strict SLA. Built high-volume scoring pipelines integrating Snowflake, Kafka, and mainframe systems. Designed optimized SQL and Snowflake queries, and deployed end-to-end ML models using Python, Domino, and OpenShift to support life-event prediction for marketing/personalization, delivering 3+ models with measurable engagement uplift. Implemented validation checks, audience validation, and model health monitoring (CSI/PSI) to ensure reliability. Enabled real-time scoring outputs via Kafka and managed secure data transfers, logging, and monitoring with Tableau dashboards.

AI/ML Engineer at TCS-USAA

May 1, 2023 - November 1, 2024

Architected and deployed a production-ready Retrieval-Augmented Generation (RAG) system for document management and automated log analysis. Built a scalable FastAPI service with asynchronous CRUD, implemented semantic search with ChromaDB and dense vector retrieval, and integrated LLaMA LLM with LangChain for contextual responses. Utilized RecursiveCharacterTextSplitter and GPT4AllEmbeddings for efficient chunking and high-quality vectorization, enabling intelligent error pattern detection and severity assessment. Unified structured and unstructured data handling via a single RESTful API interface.

ML Engineer at TCS

September 1, 2020 - August 1, 2022

Improved fraud detection precision by 18% by training supervised and transformer-based models on high-volume transaction data. Reduced model retraining time by 42% by restructuring Airflow workflows for feature generation, validation, and deployments. Increased anomaly recall by 21% using LSTM models to model sequence-level spending behavior. Reduced inference latency from 220ms to 90ms by optimizing FastAPI-based inference services on AWS. Generated SHAP explanations for scored transactions to support regulatory audits and improved feature pipeline reliability from 87% to 98% by redesigning S3-based data workflows.

ML Intern at Techimax IT Services Pvt Ltd

May 1, 2019 - June 1, 2020

Improved forecasting accuracy by 15% for retail and banking clients by tuning regression/classification models. Increased customer segmentation effectiveness by 26% using K-Means and DBSCAN. Boosted defect detection accuracy by 19% with CNN-based models, and reduced deployment cycles by 40% by containerizing ML workflows with Docker and exposing endpoints via Flask/FastAPI.