Breakfast Roundtable: AI Runs on Networks: Rethinking Enterprise Network Infrastructure for the AI Era with Tata Communications
We are currently accepting applications for this Roundtable. If you are a CIO, CTO, Heads/VPs of network Infrastructure, Sr Manager, or Director of Network Infrastructure, apply to join here: https://share.hsforms.com/1cmAOVo8hTW-CKdWr6pehLA3kiev
As enterprises operationalize AI—from generative copilots to real-time analytics and autonomous workflows—the underlying network is emerging as a critical enabler of AI performance, scale, and security. AI workloads are distributed across hybrid and multi-cloud environments, rely on massive data movement, and demand ultra-low latency connectivity between data, compute, and users. Traditional network architectures, designed primarily for predictable application traffic, are increasingly being stretched by these new patterns.
This breakfast roundtable will explore how enterprise network infrastructure must evolve to support AI-driven organizations. The discussion will examine the architectural implications of AI workloads on WAN, cloud connectivity, and edge infrastructure, as well as operational considerations such as observability, security, and automation.
Participants will also reflect on how CIOs and network leaders are aligning network strategy with AI adoption roadmaps—ensuring the network becomes a strategic platform for innovation rather than a bottleneck. The session aims to exchange practical perspectives on preparing enterprise networks for the scale, speed, and intelligence required in the AI era.
Discussion Points
1. AI Workload Architecture: What Changes for the Network?
- AI applications move large volumes of data between data lakes, training clusters, inference engines, and users across multiple clouds and regions.
- How are AI workloads changing traffic patterns compared to traditional enterprise apps?
- What new requirements are emerging for bandwidth, latency, and inter-cloud connectivity?
2. The Network as the AI Data Pipeline
- AI performance often depends less on compute and more on how quickly data can move across environments.
- How are enterprises thinking about WAN, backbone connectivity, and cloud interconnects as “data pipelines” for AI?
3. Multi-Cloud, Sovereignty, and the AI Geography Problem
- AI workloads often run across multiple regions due to regulatory, cost, or compute availability constraints.
- How are enterprises addressing connectivity challenges between regions, regulated markets, and hyperscalers?
4. AI for the Network vs. Network for AI
- AI is not only consuming networks—it is also transforming network operations.
- How are enterprises applying AI for network observability, traffic optimization, and incident response?
Raj Purkayastha is a seasoned technology and telecom leader with more than 25 years of experience spanning pre sales, business development, marketing, and delivery across enterprise and international telecom markets in both mature and emerging regions. As Vice President at Tata Communications, he leads the Pre Sales (Sales Engineering) organization for the Americas, partnering with large enterprises to shape their innovation and growth strategies. In this role, Raj works closely with CIOs and technology leaders to co create solutions that modernize digital infrastructure, simplify complexity, and accelerate what comes next for their businesses.
Throughout his tenure at Tata Communications, Raj has held multiple leadership roles across regions and business lines including Global Carrier Solutions and Enterprise Solutions, consistently delivering revenue growth, profitability, and top tier employee engagement results. He previously led India Pre Sales and served as the Global Head of Product Marketing where he helped translate complex technologies into compelling stories and customer facing narratives that drove product adoption and strengthened the company’s market presence.
Raj is deeply passionate about innovation and has expertise across technologies such as drone navigation, robotic process automation, and Internet systems. He is a patent holder with additional patents pending, and his research has been published across several emerging technology domains. At Tata Communications, he spearheaded the development of the company’s Work from Home solution stack during the COVID 19 pandemic, using internal IP to rapidly support customers’ evolving digital workspace needs. He has also led key digital transformation initiatives in retail, manufacturing, and other industries, helping enterprises adopt agile, resilient, and future ready architectures. Raj’s global impact includes leading a major country wide telecom transformation program for a government in Southern Africa that significantly expanded national network reach, improved affordability, and strengthened digital capability. His customer engagement philosophy focuses on delivering strategic points of view and thought leadership that help organizations rethink what is possible and reinforce Tata Communications’ leadership in critical markets.
Additionally, Raj has filed 6 Patents in areas of AI, Edge, Drones, RPA and Internet. 3 of these are already awarded to him.
Nick Lippis is an authority on corporate computer networking. He has designed some for the largest computer networks in the world. He has advised many Global 2000 firms on network strategy, architecture, equipment, services and implementation including Hughes Aerospace, Barclays Bank, Kaiser Permanente, Eastman Kodak Company, Federal Deposit Insurance Corporation (FDIC), Liberty Mutual, Schering-Plough, Sprint, WorldCom, Cisco Systems, Nortel Networks and a wide range of other equipment suppliers and service providers.
Mr. Lippis is uniquely positioned to comment, analyze and observe computer networking industry trends and developments.
Tony Farinacci recently retired from JPMorgan Chase after 22 years of service in executive leadership roles including CTO – Global Network Services, End-User Services, CISO – Global Technology Infrastructure, and Distinguished Architect – Cybersecurity Architecture. His first 13 years were providing global oversight of JPMorgan Chase’s Voice & Data Network, and Perimeter Security with the last 9 years in Cybersecurity focused on Infrastructure and Cloud Security, Threat Modeling, Risk Assessment, Policy as Code, and Zero Trust.
Previous to JPMorgan Chase, Tony spent 9 years at KeyCorp in various executive leadership roles including oversight of Voice & Data Networks, Enterprise Architecture, Distributed Computing, and the Enterprise Command Center.
Tony started his career at General Electric where he spent 7 years in various technical leadership roles including Senior Architect & Systems Programmer (Mainframe, UNIX, VAX, LAN), Telecommunications and Networking.
Tony earned his Bachelor of Electrical Engineering degree at Cleveland State University. He currently resides in Boca Raton, Florida with his wife.
