MCP and the future of AI integrations | Interview with Ray Bogman
Insights on MCP and AI connectivity with Ray Bogman, Head of Innovation, Alumio
Developed in 2024 by Anthropic, creators of the Claude LLM, the Model Context Protocol or MCP is rapidly emerging as a foundational standard for connecting AI assistants and LLM tools with thousands of business applications and data sources. MCP provides a standardized, model-agnostic interface that any AI assistant or tool can use to interact with a wide variety of applications and datasets, without creating bespoke integrations. As such, it’s being described as the USB-C for AI.
To understand the real-world applications of MCP better, we interviewed our resident AI expert and Head of Innovation, Ray Bogman, to give us a more grounded perspective on this new universal standard for AI connectivity. Here’s what he had to say:
What is MCP, and how is it impacting AI integrations?
“MCP, or Model Context Protocol, is like a plug-and-play solution that standardizes the way AI models, especially LLMs, can retrieve and augment data from external applications and data sources. It’s very similar to what we do with the Alumio iPaaS— bridging systems and data. The impact of MCP in the world of AI has been quite significant, since Anthropic introduced it in 2024. It doesn’t just impact how AI functions; it changes how integration is approached entirely. It’s not just an AI advancement—it’s a new way of thinking about connectivity.”
How can Alumio iPaaS users effectively leverage MCP?
“One of the key advantages of MCP is how it allows AI models to directly interact with business systems in a context-aware way. With Alumio acting as the integration layer, this becomes incredibly powerful for our customers.
For example, let’s take our customers who use the Alumio iPaaS to connect to Magento (now known as Adobe Commerce). In the near future, we envision offering an MCP server connection specifically for Magento within the Alumio ecosystem. What this means is that an AI assistant, through MCP, could query data from Magento using simple prompts. For instance, a user might ask, “What T-shirts are available in the summer collection?” or “Which SKUs belong to our ‘essentials’ category?”. With MCP in place, those prompts can trigger real-time data retrieval from Magento via Alumio, without the need for custom queries or additional configurations.
This turns what used to be complex data lookups into natural, AI-driven interactions. Alumio sees it as an opportunity to enable our customers to use AI not just for content or automation, but to dynamically augment their product catalogs, order systems, or any connected app, directly from within Alumio. It’s a smarter, more flexible way to use AI across the entire integration landscape.”
What are some other use cases for MCP?
“There are quite a few promising use cases of MCP, the potential of each depending on the availability of existing Connectors and how an application exposes its data. While Magento (Adobe Commerce) is one example, we also work with applications like Spryker, Shopify, Shopware, BigCommerce, and many others, each of which could support MCP-enabled connections through the Alumio iPaaS.
The beauty of MCP is that it builds on top of what’s already there. For instance, with Magento, we’re not asking customers to modify their systems or expose new endpoints. We’re simply using the standard REST APIs that already exist. Alumio helps configure all the intelligence and logic, such as the context formatting and model communication, which is required for the AI to understand and access the data via MCP. In other words, by leveraging basic APIs, it gives customers the ability to query the data that’s already connected to their Alumio routes.
That means the same principle can apply across other platforms too. If a business is using Shopify, for example, they could prompt an AI assistant to pull real-time stock levels, fetch order history, or even generate a sales summary, without needing to create new integrations. It’s about unlocking access to existing business data in a way that’s prompt-ready and AI-compatible.
Ultimately, MCP allows Alumio to serve as the bridge, so any system we connect to today via APIs can become AI-accessible tomorrow, with minimal friction. That’s what makes it so scalable.”