Joyjit Roy
Principal Technical Program Manager, KForce
Austin, Texas, United States
Actions
Joyjit Roy is a recognized leader in AI, Machine Learning, and Information Technology, with expertise in cloud-native platforms and large-scale digital systems. With over two decades of experience in the IT industry, he has led digital transformation initiatives across insurance, financial services, and global eCommerce sectors, including Fortune 500 companies.
His research interests focus on enterprise AI systems, Agentic AI architectures, intelligent automation, and distributed cloud platforms supporting production-scale digital ecosystems. Roy contributes to the community through research, peer review, mentorship, and technical leadership.
He frequently speaks at AI, distributed systems, and enterprise technology transformation conferences and summits.
Links
Area of Expertise
Topics
Using AI Tools to Support Special Education Planning and Administrative Workflows
Artificial intelligence tools are beginning to assist educators and administrators with documentation, planning, and workflow management. This session explores practical ways AI can support special education teams by helping streamline paperwork, organize information, and improve communication while maintaining responsible and ethical use. Participants will see examples of how AI-assisted workflows can reduce administrative burden and support more efficient program coordination.
Real-Time AI Systems with Apache Ignite: Architecting Distributed Intelligence at Scale
Modern AI systems increasingly require real-time decision making, low-latency data access, and the ability to scale across distributed environments. In-memory computing platforms such as Apache Ignite provide a powerful foundation for building intelligent, high-volume applications that demand speed and resilience.
This session explores how distributed data grids can power next-generation AI systems by enabling real-time feature access, low-latency model inference, and scalable data pipelines. The talk will discuss architectural patterns for combining Apache Ignite with cloud-native platforms, AI pipelines, and event-driven systems.
Attendees will gain insights into building resilient data architectures that support large-scale AI workloads while maintaining performance, scalability, and operational reliability.
Orchestrating Enterprise AI Pipelines with Apache Airflow
As organizations deploy machine learning and AI systems at scale, managing complex data pipelines, model training workflows, and automated decision systems becomes increasingly challenging. Apache Airflow has emerged as a powerful orchestration platform for coordinating data engineering, analytics, and AI workflows across distributed cloud environments.
This session explores how Airflow can orchestrate enterprise AI pipelines, from data ingestion and feature engineering to model training, evaluation, and deployment. Drawing from real-world enterprise transformation programs, the talk will highlight architectural patterns for integrating Airflow with modern AI platforms, cloud-native infrastructure, and governance frameworks.
Attendees will gain practical insights into designing reliable AI workflows, managing dependencies across complex data pipelines, and scaling orchestration for production-grade machine learning systems.
From Legacy to Cloud-Native: Modernizing a Policy Platform on Kubernetes at Enterprise Scale
This session shares lessons from modernizing a large, regulated enterprise platform from a legacy, tightly coupled architecture to a Kubernetes-based cloud-native environment. The journey involved decomposing monolithic services, introducing containerized workloads, and establishing a platform foundation that balanced developer autonomy with enterprise governance, security controls, and audit requirements.
We will cover practical design decisions around cluster architecture, workload isolation, CI/CD standardization, policy enforcement, and observability at scale. The talk also discusses what did not work, including early assumptions around service boundaries, release coordination across multiple teams, and platform adoption friction.
From a business perspective, the modernization enabled faster release cycles, improved platform reliability, better resource utilization, and reduced operational risk in a compliance-heavy domain. Attendees will gain concrete patterns, architectural tradeoffs, and operational lessons they can apply when modernizing critical enterprise systems on Kubernetes.
Coordination Patterns for Multi-Agent AI Systems in Enterprise Production
Modern enterprise AI systems are evolving from isolated AI agents toward coordinated multi-agent architectures capable of adaptive reasoning, distributed decisions, and autonomous execution.
This session covers coordination patterns in production-grade multi-agent AI systems, including task decomposition, agent orchestration, shared memory layers, feedback loops, and risk-aware decision boundaries. It also examines how specialized agents collaborate across enterprise workflows while maintaining governance and human oversight.
Drawing on enterprise deployment examples, the session highlights how multi-agent systems improve operational efficiency, accelerate decision-making, and enable scalable autonomous operations. It also addresses architectural considerations for deploying agentic systems in production, including orchestration frameworks, communication protocols, runtime coordination, monitoring, and safety guardrails.
Attendees will gain a practical understanding of how coordinated multi-agent systems differ from standalone AI agents, emerging architectural patterns in enterprise deployments, and the challenges organizations face when operationalizing agentic AI at scale.
Building Autonomous Edge AI Systems: From Architecture to Production at Scale
As AI workloads increasingly move from centralized cloud environments to edge infrastructure, organizations face new challenges in deploying intelligent systems across large fleets of distributed devices. This session explores architectural patterns for building autonomous Edge AI systems that combine local decision-making with cloud orchestration.
Drawing on real-world experience deploying Edge AI platforms across IoT and enterprise environments, the talk covers distributed AI architectures, agentic decision systems at the edge, and orchestration strategies using cloud-native technologies. Attendees will gain practical insights into designing scalable, resilient Edge AI platforms capable of operating reliably across millions of devices.
AgentSOC: Autonomous Agentic AI for Real-Time Cybersecurity Operations
Cyberattacks cost the global economy an estimated $10.5 trillion annually. Enterprise SOCs receive over 100,000 alerts daily, 70% go uninvestigated, and 3.5 million security positions remain unfilled worldwide. Attackers exploit every one of those gaps.
Traditional platforms match known patterns and run fixed playbooks. They cannot predict what an attacker will do next, verify if that prediction holds in your specific environment, or assess whether a defensive action causes more damage than the attack itself. The result is a slow, inconsistent response at exactly the wrong moment.
This session presents AgentSOC, a framework that closes all three gaps in a single autonomous cycle. It reasons across multiple attack hypotheses, validates each against real network topology and access structure, and selects the response that best balances threat containment with business impact.
Attendees will walk away with a concrete understanding of where current automation fails, how agentic reasoning changes the defender equation, and what responsible closed-loop SOC design looks like in practice.
From Copilot to Autonomous Workflows: The Rise of Agentic AI in the Enterprise
Enterprise AI is rapidly evolving beyond copilots that assist users toward agentic systems capable of autonomously executing tasks, coordinating workflows, and adapting to changing business conditions. As organizations adopt Microsoft Copilot and AI-powered platforms, the next challenge is extending these capabilities into secure, governance-aware automation across the enterprise.
In this session, we explore how agentic AI systems can build on the Microsoft ecosystem to enable intelligent workflow orchestration, operational automation, and scalable enterprise decision support. The discussion will cover architectural patterns, governance considerations, and practical lessons from deploying AI-driven systems in regulated industries.
Attendees will gain a practical understanding of how organizations can transition from AI-assisted workflows to autonomous enterprise workflows while maintaining security, compliance, and operational control.
Key Takeaways:
1. Understand the evolution from Copilot-style assistance to agentic AI systems
2. Learn architectural patterns for enterprise AI automation within modern cloud ecosystems
3. Explore governance, security, and risk considerations when deploying autonomous AI workflows in production environments
From AI Agents to Agentic Systems: Architecting Autonomous AI for Production
This session examines the architectural difference between simple task-based AI agents and fully agentic systems capable of adaptive planning, autonomous execution, and continuous feedback-driven refinement. Rather than focusing on hype, the discussion centers on practical implementation patterns observed across eCommerce, insurance, and healthcare environments.
Core system components are explored, including intent interpretation, adaptive planning loops, guardrails, execution controllers, observability layers, and human-in-the-loop safeguards. Real-world implementations such as AI-powered underwriting platforms, conversational shopping agents, and preoperative healthcare assistants demonstrate how orchestration frameworks replace rigid workflow automation.
Attendees will gain a clear architectural perspective on evolving from prompt-driven automation toward scalable, governable, multi-agent production systems.
AI-Driven Security as a Competitive Advantage in Regulated Industries
AI adoption in cybersecurity is accelerating, but many organizations struggle to translate technical innovation into clear business value. Based on real-world enterprise deployments across regulated industries, this session explores how AI-driven security automation improves resilience, strengthens brand trust, and supports executive-level risk narratives.
In this session, we will examine how agentic AI frameworks in modern SOC environments affect customer confidence, operational maturity, and strategic differentiation. Attendees will gain practical insights for aligning AI-powered security initiatives with marketing, trust, and growth objectives.
Modernizing a Legacy Policy Platform on Kubernetes: Architecture, Tradeoffs, and Enterprise Outcomes
This session shares lessons from modernizing a large, regulated enterprise platform from a legacy, tightly coupled architecture to a Kubernetes-based cloud-native environment. The journey involved decomposing monolithic services, introducing containerized workloads, and establishing a platform foundation that balanced developer autonomy with enterprise governance, security controls, and audit requirements.
We will cover practical design decisions around cluster architecture, workload isolation, CI/CD standardization, policy enforcement, and observability at scale. The talk also discusses what did not work, including early assumptions about service boundaries, release coordination across multiple teams, and friction with platform adoption.
From a business perspective, the modernization enabled faster release cycles, improved platform reliability, better resource utilization, and reduced operational risk in a compliance-heavy domain. Attendees will gain concrete patterns, architectural tradeoffs, and operational lessons they can apply when modernizing critical enterprise systems on Kubernetes.
Building Autonomous Edge AI Systems: From Architecture to Production at Scale
As AI workloads increasingly move from the cloud to the edge, organizations face critical challenges in orchestrating intelligent systems across millions of heterogeneous devices. This session explores the architectural patterns and practical strategies for building production-grade Edge AI platforms that operate autonomously while maintaining coordination with cloud-based orchestrators.
Drawing from real-world experience deploying AI systems across industrial IoT, healthcare, and smart infrastructure environments, this talk covers:
Edge AI Architecture Patterns: Designing resilient, distributed systems that leverage CPU/GPU/TPU/NPU heterogeneity
Agentic AI at the Edge: Building multi-agent systems that make autonomous decisions with minimal latency
Orchestration Strategies: Balancing edge autonomy with cloud coordination using modern messaging systems (gRPC, MQTT) and cloud-native technologies (Kubernetes, Docker)
Production Lessons: Key insights from scaling Edge AI platforms to millions of devices.
Enterprise Scale eCommerce Modernization
This session unpacks the strategic and operational lessons learned from a large eCommerce Service's modernization initiative, detailing the shift from a traditional SKU-based model to a unified, subscription-driven ecosystem. Attendees will learn how vision, architecture, governance, and leadership aligned to deliver one of the organization’s largest transformations, providing practical insights for digital leaders and program managers on scaling Agile delivery and ensuring sustainable modernization.
Agentic AI for Business Transformation
This strategic roadmap defines the next era of Agentic AI, a critical shift beyond automation. We detail leveraging AI Agents and Multi-Agent Systems to perceive, reason, and execute complex, multi-step tasks across enterprise systems. The scope covers intelligent program orchestration, optimizing operational efficiency, and building adaptive, resilient, cloud-native ecosystems that secure competitive advantage and deliver measurable outcomes globally.
DSS ATX 2026
Data Science Salon is a flagship Austin, TX event hosted at Oracle HQ on February 18, 2026. This intimate, one-day conference brings together approximately 250 attendees and curates high-quality data science sessions that connect industry leaders and specialists to share innovative solutions in generative AI, machine learning, predictive analytics, and best practices. The event features a mix of use cases, technical talks, and workshops, enabling attendees to gain actionable insights from professionals working on the front lines of enterprise machine learning.
The session I presented explored how Agentic AI architectures move beyond rule-based workflows to enable enterprise-scale decision automation through adaptive planning, contextual memory, and closed-loop evaluation. It outlined key architectural components including guardrails, task decomposition, execution control, and feedback-driven refinement, and demonstrated how such systems maintain auditability while operating under uncertainty. Real-world examples from eCommerce, insurance, and healthcare highlighted measurable improvements in personalization, underwriting, fraud detection, and clinical coordination, with a strong emphasis on governance, human oversight, and production-ready deployment.
Austin ACM/IEEE Meetup - TechTalks User group Sessionize Event
ProjectManagement.com Webinar Series: Enterprise Scale eCommerce Modernization
This webinar presented a real-world enterprise transformation from a traditional SKU-based commerce model to a subscription-centric digital ecosystem. The session covered cloud-native architectural evolution, scaled Agile delivery, cross-functional coordination, governance at scale, and leadership decision-making in complex modernization programs.
The talk focused on how technical architecture, platform strategy, and execution frameworks intersect in large enterprise environments. Practical lessons included risk management, alignment across engineering and business teams, and sustaining long-term value in subscription-based platforms.
The session was hosted on ProjectManagement.com (PMI platform) and qualifies for 1.0 PDU for PMI-certified professionals.
Enterprise Modernization: Mastering the Shift to Subscription and Continuous Value Models
Session Type: Technical Webinar
Platform: Luma (hosted via Google Meet)
Presenter: Joyjit Roy
Delivered a professional webinar focused on enterprise modernization and the transition from traditional product models to subscription and continuous value delivery. The session covered cloud-native engineering approaches, modernization frameworks, platform evolution, and scaled Agile execution in enterprise environments.
The talk emphasized architectural strategy, leadership alignment, and practical lessons from large-scale digital transformation initiatives. The event was publicly listed and attended by technology and engineering professionals interested in modernization and platform transformation topics.
A recording of the session and supporting materials are available.
2nd International Conference on Sustainable Computing and Intelligent Systems (SCIS 2025)
Session Type: Invited Conference Presentation (Virtual)
Event: SCIS 2025 – 2nd International Conference on Sustainable Computing and Intelligent Systems
Organizer: Sustainable Computing and Intelligent Systems (SICS)
Delivered a technical session at SCIS 2025 focused on intelligent systems and applied computing within sustainable and enterprise-scale technology environments. The talk covered system design considerations, architectural tradeoffs, and applied AI/ML integration in complex computing ecosystems.
The conference was hosted by the University of Canberra, Australia, and brought together researchers and practitioners from sustainable computing, intelligent systems, and applied engineering domains. The presentation was delivered virtually.
Speaker profile, session record, and participation certificate are available via the conference portal.
Austin Kubernetes Meetup User group Sessionize Event
Joyjit Roy
Principal Technical Program Manager, KForce
Austin, Texas, United States
Links
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top