The voice of AI, Data and Innovation
About the Guest
Romain Sestier
Romain Sestier is the Co-founder and CEO of StackOne, a company building integration infrastructure that enables AI agents to securely connect with and act across enterprise applications.
With over a decade of experience building and scaling SaaS products, Romain previously held leadership roles at Google, including its Area 120 innovation lab, Yieldify, and Contentsquare. His experience with complex enterprise integrations led him to co-found StackOne, which raised a $20 million Series A led by GV (Google Ventures), with participation from Workday Ventures and other investors.
Today, he focuses on solving the infrastructure challenges behind enterprise AI adoption, particularly agent connectivity, security, governance, and tool execution.
Brief About the Episode
Connecting an AI agent to enterprise applications is one thing. Allowing it to securely access sensitive data, choose the right tools, and take actions across systems is an entirely different challenge.
In this episode, Jas Kaur sits down with Romain Sestier, Co-founder and CEO of StackOne, to explore why traditional APIs and the Model Context Protocol (MCP) alone aren’t enough to support production-ready enterprise AI agents.
The conversation examines the growing need for tool gateway, an infrastructure layer that helps agents discover tools, execute actions, enforce permissions, and interact securely with business applications.
From agent security and prompt injection risks to context window limitations, token optimization, and granular access controls, Romain shares how enterprises can move beyond basic connectivity toward reliable, governed agentic workflows.
Key Learnings for Leaders
✦
AI agents need more than API and MCP connectivity
Traditional APIs were designed for developers executing predefined requests, not autonomous agents for selecting actions dynamically. MCP helps standardize connectivity, but enterprises still need governance, security, and intelligent tool discovery to manage complex agent workflows.
✦
Tool gateways are becoming critical infrastructure.
An AI agent gateway sits between AI systems and enterprise applications, helping agents discover relevant tools, execute authorized actions, and enforce security policies. This layer becomes increasingly important as organizations connect agents to hundreds of tools and business systems.
✦
Enterprise AI governance must scale beyond individual agents
As organizations deploy hundreds or thousands of AI agents, securing each agent independently becomes increasingly difficult. Centralized governance, monitoring, and consistent access policies are essential to scaling agentic AI without sacrificing business security.
Watch the episode to discover why APIs and MCP alone aren’t enough and what it takes to build AI agent infrastructure ready for enterprise-scale deployment.
Join our mailing list
Be the first to know when new episodes drop.
Subscribe to get fresh drops, exclusive guest insights, and key takeaways in your inbox.
