Enterprise Data & AI leadership for agentic systems, governance, and measurable impact.
Prince Paulraj is an executive Data, AI, and Agentic Systems leader at AT&T, based in Dallas, Texas, with a proven record of translating advanced AI into enterprise-scale business value. He leads high-impact AI initiatives across Finance, AT&T Business, and customer platforms, driving revenue growth, cost optimization, and risk reduction through production-grade systems. Previously, he served as Chief Data Officer (India), where he built AT&T's Chief Data Office from inception and shaped enterprise data and AI strategy. His work includes billions in fraud loss prevention, enterprise GenAI platforms, and foundational MLOps infrastructure.
Impact
Outcome-driven AI: fraud prevention, operational productivity, optimization, and finance transformation.
GenAI-Powered Platforms
Launched enterprise GenAI platforms including Ask AT&T, Ask Ops, AI Agent Foundry, Summarise AI, OpsGPT, and FraudGPT, enabling multi-agent workflows, LLM-driven systems, and RAG-powered solutions that transform operations and customer experiences. These platforms have significantly improved AT&T's operational efficiency and profitability.
Fraud Detection & Prevention
Developed AT&T's fraud detection systems that have prevented billions in transaction fraud, making substantial earnings per share (EPS) contributions using traditional machine learning and AI technologies.
ML Feature Store & Platform
Co-invented and co-developed the H2O AI Feature Store with H2O.ai, enabling production-grade feature reuse for large-scale and real-time workloads. Created Watchtower for enterprise-wide AI governance, monitoring, and scalable adoption. Delivered measurable improvements in model development velocity and production reliability.
Expertise
Strategy-to-execution leadership across data, AI, and modern engineering practices for enterprise scale.
Generative AI & LLM Systems
Building production GenAI platforms with multi-agent workflows, RAG architectures, and LLM-driven solutions for enterprise use cases including digital receptionists, knowledge retrieval, and autonomous operations.
ML Feature Store & MLOps
Co-invented and co-developed H2O AI Feature Store with H2O.ai, establishing production-grade feature reuse infrastructure. Enables governed delivery, versioning, and scalable AI adoption across teams through AT&T's AI-as-a-Service (AIaaS) platform. Built Watchtower for enterprise-wide AI governance and monitoring.
Fraud Detection & Risk Management
Enterprise-scale fraud detection systems using machine learning and AI, preventing billions in transaction fraud with real-time monitoring and risk assessment.
Data Engineering & Big Data
Leading teams of 100+ data engineers and scientists, building forecasting models, real-time network monitoring pipelines, and cost-saving data insights at scale.
Career timeline
Leadership across telecom, marketplaces, and global delivery—building systems from engineering foundations to enterprise-scale AI.
Patents & Publications
Innovation in AI, machine learning, fraud detection, and telecommunications. Key invention themes include AI platform governance, telecom intelligence, and robust production ML. View on Google Scholar
117 Citations
Total citations across all publications, with 106 citations since 2020, demonstrating sustained impact and relevance in AI, machine learning, and telecommunications research.
Research Quality
h-index of 5 indicates consistent quality—five publications each cited at least five times. This metric reflects both productivity and citation impact in the field.
20+ Patents
Granted and filed patents covering fraud detection, network optimization, machine learning infrastructure, and AI governance systems.
Video pin sharing
US Patent 9,653,116 (2017). Co-invented with D Srivastava. System for sharing video content through pin-based mechanisms.
System and method to identify failed points of network impacts in real time
US Patent App. 15/986,324 (2019). Co-invented with L Haugen, C Tsai, H Miao, P Gururaj, S Harpavat, S Meredith. Real-time network failure detection and impact analysis system.
Telecommunication network machine learning data source fault detection and mitigation
US Patent 20,220,329,328 (2022). Co-invented with A Armenta, L Savage. ML-based system for detecting and mitigating data source faults in telecommunications networks.
Relationship graphs for telecommunication network fraud detection
US Patent 12,192,400 (2025). Co-invented with S Murali, EJ Abrahamian, A Armenta, E Hall. Graph-based approach to detecting fraud patterns in telecommunications networks.
Transformation as a Service
US Patent 20,220,318,194 (2022). Co-invented with P Ireifej, MOK Mirza, H Wighton, C Kim, S Grandinetti. Service-oriented architecture for enterprise transformation capabilities.
Data stream based event sequence anomaly detection for mobility customer fraud analysis
US Patent 11,979,521 (2024). Co-invented with R Steckel, A Armenta, CC Huang. Real-time anomaly detection system for identifying fraudulent patterns in mobility customer data streams.
Data harmonization across multiple sources
US Patent 11,625,379 (2023). Co-invented with S Harpavat, W Liu, S Taywade, AC Nagarasan, Y Zeng. System for harmonizing and integrating data from multiple heterogeneous sources.
Governance mechanisms for reuse of machine learning models and features
US Patent 12,481,915 (2025). Co-invented with C Kim, E Zavesky, P Sugumaran, J Pratt, C Vo. Framework for governing and enabling safe reuse of ML models and features across the enterprise.
Machine learning model feature sharing for subscriber identity module hijack prevention
US Patent 12,363,087 (2025). Co-invented with A Diffloth, J Pratt. ML-based system for detecting and preventing SIM hijacking through feature sharing mechanisms.
Steering of roaming optimization with subscriber behavior prediction
US Patent App. 17/331,225 (2022). Co-invented with Y Zeng, S Rogers, S Alexander, S Harpavat, S Taywade. ML-driven system for optimizing roaming services based on subscriber behavior predictions.
Sensory density and diversity for living in place
US Patent 20,170,046,497 (2017). Co-invented with Vc Ramesh, Michael G. Branam, Philip Edward Brown, Lee. System for analyzing sensory data to support aging in place.
Trust labeling of call graphs for telecommunication network activity detection
US Patent 12,301,425 (2025). Co-invented with E Hall, A Armenta. Trust-based labeling system for call graphs to detect suspicious network activities.
Restricted reuse of machine learning model data features
US Patent App. 17/949,787 (2024). Co-invented with A Diffloth, J Pratt. Governance framework for controlling and restricting reuse of ML model features.
Code-to-utilization metric based code architecture adaptation
US Patent App. 17/520,144 (2023). Co-invented with A Campbell, S Taywade. System for adapting code architecture based on utilization metrics.
Machine learning feature recommender
US Patent 20,220,327,401 (2022). Co-invented with PP Joshua Whitney, Edmond J. Abrahamian. AI-powered system for recommending relevant ML features for model development.
Call graphs for telecommunication network activity detection
US Patent 11,943,386 (2024). Co-invented with E Hall, A Armenta, S Murali. Graph-based system for detecting suspicious activities in telecommunications networks.
Similarity-based search for fraud prevention
US Patent App. 17/382,746 (2023). Co-invented with A Luthra, A Armenta, J Luo. Similarity-based search algorithms for detecting and preventing fraudulent activities.
Anomaly detection relating to communications using information embedding
US Patent 12,470,569 (2025). Co-invented with EJ Abrahamian, A Campbell, A Armenta. Information embedding techniques for detecting anomalies in communication networks.
Machine learning model feature sharing for subscriber identity module hijack prevention
US Patent App. 19/268,753 (2025). Co-invented with A Diffloth, J Pratt. Enhanced ML-based system for preventing SIM hijacking through feature sharing.
Mitigating temporal generalization for a machine learning model
US Patent App. 19/252,218 (2025). Co-invented with BB Lee, A Campbell, A Armenta. Techniques for addressing temporal generalization challenges in ML models.
Awards & Recognition
Industry recognition for leadership in AI, data science, and enterprise innovation.
Award Badges
Award Timeline
H2O.ai Top 100 AI Leaders 2025
H2O.ai • Innovators - Enterprise Category
2025Recognized as a Top 100 AI Leader in the Innovators - Enterprise category for driving real-world AI impact. Honored alongside leaders from Dell, NVIDIA, eBay, and other Fortune 500 companies for translating AI from promise into production-scale progress.
Top 100 Most Influential AI Leaders in India 2024
Analytics India Magazine (AIM)
2024Recognized as one of the 100 Most Influential AI Leaders in India for leading AT&T's Chief Data Office in India, developing GenAI products including Summarise AI, OpsGPT, and FraudGPT, and modernizing AT&T's Data and AI ecosystem through generative AI and machine learning.
MongoDB Innovation Award
MongoDB • From Batch to Real-Time Category
2022To build its next-generation AI-based fraud-detection platform, AT&T quickly discovered that relational technology would not be able to scale and support their application's needs and requirements. Given their desire for a flexible data model, AT&T turned to MongoDB Atlas, which has decreased their time to market and improved their query response times. As part of an overall modernization effort to enhance an already robust AI environment.
Speaking & media
Public sessions and case studies on AI platforms, autonomy, and operationalizing AI responsibly.
H2O.ai Top 100 AI Leaders 2025
Recognized as a Top 100 AI Leader in the Innovators - Enterprise category by H2O.ai for driving real-world AI impact. Recognized alongside leaders from Dell, NVIDIA, eBay, and other Fortune 500 companies for translating AI from promise into production-scale progress.
AIM Top 100 Most Influential AI Leaders in India 2024
Recognized as one of the 100 Most Influential AI Leaders in India 2024 by Analytics India Magazine (AIM). Honored for leading AT&T's Chief Data Office in India, developing GenAI products including Summarise AI, OpsGPT, and FraudGPT, and modernizing AT&T's Data and AI ecosystem through generative AI and machine learning.
AI Advancement using MongoDB @ AT&T
Presentation at MongoDB.local Dallas discussing AT&T's AI advancement initiatives using MongoDB. Covers use cases including Fraud.AI, H2O Feature Store integration, and real-time monitoring for AI operations. Highlights how MongoDB Atlas supports AT&T's next-generation AI-based fraud-detection platform and modern data architecture.
AI agents on the frontline: Defeating fraud & unlocking next gains
TM Forum Innovate Americas 2025. Featured speaker discussing shifting from rule-based automation to autonomous, goal-directed AI agents that self-learn, adapt, and optimize retail network and business operations. Addressing key challenges in retail stores and how Agentic AI can transform assisted customer experiences and operational efficiency, while applying TM Forum frameworks to scale Agentic AI and Autonomous Operations across Retail, RAN, and adjacent domains.
AT&T Generative AI: Empowering Employees for Innovation
Exploring how AT&T leverages generative AI technology to enhance employee effectiveness, creativity, and innovation. This talk delves into AT&T's transformational journey, where AI has been progressively integrated across the company to deliver superior value, streamline operations, unlock revenue streams, and empower employees to improve productivity and generate novel solutions.
H2O.ai Feature Store Session: AT&T Production Implementation
Joint session with Vinod Iyengar (VP of Product, H2O.ai), Prince Paulraj (AVP - Engineering, Data Science & AI, AT&T), and Jakub Hava (Lead Software Engineer, H2O.ai). Exploring how AT&T and H2O.ai jointly built an AI Feature Store to manage and reuse data and ML engineering capabilities. The Feature Store is in production at AT&T, meeting high levels of performance, reliability, and scalability.
AT&T Presents: Democratizing Data, AI & Generative AI
During the AT&T Presents technology and leadership session in Bangalore, India, Prince Paulraj shares how AT&T democratizes Data, AI, and Generative AI, practices responsible & ethical AI, and delves into the success of Ask AT&T, the company's Enterprise Secured Generative AI Platform. Featuring TED Talk style presentations, AT&T Presents provides employees with the opportunity to share their career journeys and highlights of their work in Data and AI.
Democratized AI using H2O - H2O World India
Talk by AT&T at H2O World India on democratizing AI using H2O. This presentation explores how AT&T leverages H2O.ai's platform to democratize AI capabilities across the organization, making advanced AI tools and technologies accessible to teams and enabling broader adoption of AI solutions.
AI Modernization at AT&T and Application to Fraud with Databricks
Exploring AT&T's AI modernization efforts in the cloud with Databricks and in-house developments. AT&T has been involved in AI from the beginning, with many firsts including "first to coin the term AI", "inventors of R", and "foundational work on Convolutional Neural Nets". This talk highlights the AI modernization effort and its application to Fraud, one of AT&T's biggest benefitting applications, showcasing how modern cloud infrastructure enables scalable fraud detection systems.
Education
Academic qualifications and professional development programs.
Academic Qualifications
Master's in Computer Applications from St. Joseph's College, India. Comprehensive foundation in computer science, software engineering, and information systems.
Professional Development
- Chief Data Analytical Officer Leadership Academy, Deloitte US
- Executive Program in AI for Business Leaders
Contact
For speaking, advisory, or collaboration inquiries, use the form below (connect it to your email service of choice).
Send a message
Connect & Resources
Connect on LinkedIn or explore public resources, case studies, and speaking engagements.