Agentic AI, MCPDevSummit June 9–10

- The Linux Foundation lists MCP Dev Summit Bengaluru for June 9-10, 2026, as a two-day event focused on production AI agents using Model Context Protocol. - AMD said on May 7 agentic AI can push CPU-to-GPU infrastructure closer to 1:1, a shift from earlier 1:4-8 chatbot-era designs. - June 9 sessions in Bengaluru list speakers from IBM, Google and AWS on MCP security, orchestration and cross-cloud operations.

The Linux Foundation has scheduled MCP Dev Summit Bengaluru for June 9-10, 2026, in Bengaluru, India, describing it as a gathering for developers building AI agents with the Model Context Protocol. The event page says the summit is aimed at MCP maintainers, contributors and engineers working on systems that connect models to tools, data and real-world software. The posted schedule shows two days of keynotes, breakout sessions, workshops and a solutions showcase. The event matters because the current discussion around “agentic AI” has moved beyond chatbot demos and toward production workflows. The Linux Foundation’s description frames MCP as a way to let interoperable agents securely connect to external systems, while the session list emphasizes reliability, audit trails, orchestration, tool design and security controls. That is the language of deployment, not just model capability. (events.linuxfoundation.org) ### What exactly is on the Bengaluru agenda? The June 9 schedule lists sessions including “Improving Reliability in MCP Applications Through Tool Design,” “MCP in Production: OAuth, Session Isolation, and Audit Trails for Remote Servers,” and “Designing a Control Plane for Agentic Systems Using MCP.” The same day also includes a workshop on distributed multi-agent workflows and a Google-linked lab on autonomous cross-cloud operations. (events.linuxfoundation.org) IBM appears multiple times on the posted program. The schedule lists IBM speakers on “Context-Aware MCP Servers for Small Language Models” and “Voice-First MCP: Real-Time Tool Calling Through a Spoken Interface.” Google is listed in a June 9 workshop alongside Wells Fargo, and AWS appears through Dheeraj Choudhary in an OWASP-focused security session. ### Why are developers talking about workflows instead of chat? (mcpbengaluru26.sched.com) AMD said in a May 7 blog post that agentic AI changes the infrastructure profile because an agent does more than answer a prompt. The company said agents break goals into steps, call models, query databases, connect to APIs, run enterprise applications, check permissions, retrieve memory and validate outputs in loops. (mcpbengaluru26.sched.com) That workflow description matches the MCP summit’s focus on tool access and production controls. The Linux Foundation says the Bengaluru event will cover protocol development, best practices for MCP-powered systems, and the security and operational challenges of running AI agents in production. ### Where does healthcare fit into this story? (amd.com) AWS has already been pitching agentic AI in healthcare as an operations tool. In a public-sector blog post, AWS said its healthcare agentic AI offerings can reason, plan and take autonomous actions with minimal human oversight, and pointed to uses in patient care, operational efficiency and system resilience. (events.linuxfoundation.org) Amazon also launched Amazon Connect Health this year as what it called a purpose-built AWS solution for healthcare providers. Amazon said the product is designed to reduce administrative burden and speed access to care by handling patient verification, scheduling, medical histories, documentation and coding. OpenAI, in a January 8 announcement, separately said its healthcare products are designed to reduce administrative work and support clinical workflows while meeting HIPAA-related requirements. (aws.amazon.com) ### What is the CPU-to-GPU debate developers keep citing? AMD said the old chatbot-style architecture typically used one CPU head node for a server with four to eight GPUs. In agentic AI, AMD said, the ratio is moving toward 1:1 and can be even more CPU-heavy because orchestration, tool calls, policy checks and security enforcement run outside the model itself. (aboutamazon.com) That claim has circulated widely because it shifts the hardware conversation from pure accelerator demand to the broader systems needed to run agents in production. The summit schedule reinforces that point indirectly: many of the listed talks are about transport layers, schema evolution, control planes, security models and orchestration patterns rather than benchmark scores. (amd.com) ### What happens next in Bengaluru? June 9 and June 10 are the next concrete dates. The Linux Foundation’s event page says the summit will include keynotes, breakout sessions, a solutions showcase and an attendee reception, and the public schedule says timing and room assignments remain subject to change. (events.linuxfoundation.org) (amd.com)

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.