What an MCP implementation looks like at a CRM company

Key takeaways

MCP is emerging as a crucial protocol for AI interactions and integrates well with current SaaS products.
Implementing MCP poses challenges particularly around security, permissions, and integration with existing systems.
HubSpot is experiencing a cultural shift towards AI adoption and fluency among its teams, enhancing internal workflows.
The future of MCP looks promising, with potential for further development and significant use cases across industries.

Summary

Introduction to Model Context Protocol (MCP)

In the episode, Ryan Donovan talks with Karen Aing, the EVP of Product at HubSpot, about the emerging Model Context Protocol (MCP), which facilitates communication between AI agents and various SaaS products. Karen shares her journey into the software industry, discussing how her background in programming languages has influenced her work with MCP at HubSpot.

The Emergence of MCP as a Standard

Karen highlights the significance of MCP in the realm of AI, likening its development to past web standards. She emphasizes that after OpenAI accepted MCP, significant tech players quickly followed suit. This adoption indicates that MCP could potentially become a standard protocol for AI interactions.

Challenges in MCP Implementation

The conversation delves into the challenges HubSpot faced while building their MCP server. Karen explains the intricacies of ensuring security, integrating existing systems, and managing user permissions with OAuth protocols, which were crucial for maintaining access control and data integrity.

Internal Impact of MCP at HubSpot

Karen discusses how the implementation of MCP has affected internal operations at HubSpot, fostering a culture of adopting AI tools within engineering and product teams. She mentions the high adoption rate of AI fluency among employees and how quickly these tools are being integrated into workflows.

The Future of MCP and AI Agents

Toward the end, they ponder the future of MCP and its role in the AI landscape. Karen expresses optimism about the evolving nature of MCP and its potential to unlock new scenarios. She reflects on how the advancements in AI tools are set to further enhance productivity and create new opportunities for users.

Show notes

Ryan chats with Karen Ng, EVP of Product at HubSpot, to chat about Model Context Protocol (MCP) and how they implemented it for their server for their CRM product. They chat the emergence of this as the standard for agentic interactions, the challenges of implementing the server and integrating it with their ecosystem, and how agentic AI has affected work at Hubspot.

Episode notes:

Hubspot is a customer-relationship management (CRM) platform that aims to help businesses grow.

MCP is an open-source protocol for connecting AI agents to external systems, originally developers at Anthropic.

Connect with Karen on LinkedIn.

Virtual hi-five to Rob Truxal for asking What is the purpose of "pip install --user ..."? and garnering themselves a Stellar Question badge.

See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.