Uploaded October 2025 | Updated September 2026, 2 weeks ago
Join Yann Jouanin, Director of Engineering Strategy and Transformation at TheFork and MCP Contributor, and Julien, Chief Evangelist at Arcee AI, in this in-depth discussion about MCP (Model Context Protocol): what it really is, why it matters today, and how it could reshape the way we interact with models.
⭐️⭐️⭐️ Don't forget to subscribe to be notified of future videos. You can also follow me on Medium at julsimon.medium.com or Substack at https://julsimon.substack.com. ⭐️⭐️⭐️
We start with a quick introduction of both speakers and the relevance of MCP in today’s AI ecosystem, then dive into:
- Defining MCP: what it is (and what it isn’t).
- Real capabilities: how it compares to function calling or traditional APIs, with concrete examples and anecdotes.
- Users & use cases: who should use MCP, when it’s ideal, and when it’s an overkill or even risky.
- Limits & risks: current flaws, whether they’re being addressed, and what to watch out for.
- Adoption strategy: should you wait or jump in now? Which alternatives or complementary tools exist? How to start small without unnecessary risks.
- Conclusion: Yann shares why he believes in MCP, while Julien explains why he’s skeptical but may eventually have to give in.
If you’re curious about how MCP could impact developers, companies, and the future of AI integration, this conversation is for you.
Join Yann Jouanin, Director of Engineering Strategy and Transformation at TheFork and MCP Contributor, and Julien, Chief Evangelist at Arcee AI, in this in-depth discussion about MCP (Model Context Protocol): what it really is, why it matters today, and how it could reshape the way we interact with models.
⭐️⭐️⭐️ Don't forget to subscribe to be notified of future videos. You can also follow me on Medium at julsimon.medium.com or Substack at https://julsimon.substack.com. ⭐️⭐️⭐️
We start with a quick introduction of both speakers and the relevance of MCP in today’s AI ecosystem, then dive into:
- Defining MCP: what it is (and what it isn’t).
- Real capabilities: how it compares to function calling or traditional APIs, with concrete examples and anecdotes.
- Users & use cases: who should use MCP, when it’s ideal, and when it’s an overkill or even risky.
- Limits & risks: current flaws, whether they’re being addressed, and what to watch out for.
- Adoption strategy: should you wait or jump in now? Which alternatives or complementary tools exist? How to start small without unnecessary risks.
- Conclusion: Yann shares why he believes in MCP, while Julien explains why he’s skeptical but may eventually have to give in.
If you’re curious about how MCP could impact developers, companies, and the future of AI integration, this conversation is for you.
