Chloe Maraina is a force in the world of business intelligence, known for her ability to transform vast, complex datasets into clear, actionable narratives. As a leading BI expert with a deep passion for data science and a forward-looking vision for data management, she has a unique perspective on the challenges and opportunities facing enterprises today. We’ll explore the critical gap between AI experimentation and production-ready applications, discussing how to overcome the common hurdles of governance and security that derail so many projects. We will also delve into the power of integrated platforms, the potential for autonomous AI agents, and the future of cross-platform AI collaboration.
Many initial AI pilots start with what’s being called “vibe coding,” but they often hit a wall when it comes to production because of fragile code and compliance issues. How does a tool like App Catalyst bridge this gap, and could you walk us through how a developer might take a simple idea and build a genuine, enterprise-grade app with it?
That’s the central challenge we’re seeing everywhere. A developer gets a great idea, uses a natural language tool to quickly generate some code, and creates an exciting prototype. It feels like magic. But that magic fades fast when you try to connect it to real, governed enterprise data. The code is often brittle and completely disconnected from security protocols. App Catalyst changes that entire dynamic. A developer can still start with that same simple idea, using a natural language prompt to get started. The difference is, the code generated is built on Domo’s foundation. This means from the very first line, it’s already integrated with our existing data governance frameworks and security permissions. You’re not just getting raw code; you’re getting a compliant, secure scaffold for your application, turning what used to be a fun side project into a legitimate business app without hitting those frustrating deployment roadblocks.
It sounds like integrating governance and security from the beginning is a core part of the solution. Can you elaborate on how App Catalyst manages to bake these standards into the development process so early, and what specific hurdles this approach helps data and AI teams leap over?
Absolutely. This is probably the most significant value it offers. Traditionally, governance and security are afterthoughts. A team builds a brilliant AI tool, and then they take it to the security team, who immediately puts the brakes on because it doesn’t meet any of the compliance standards. That’s how projects die. What we’re doing is baking it in from the start. As soon as a developer begins building with App Catalyst, the tool is already aware of the entire governance layer within the Domo platform. It automatically integrates with established security permissions and data access policies. This completely eliminates that “compliance headache” that kills so many initiatives. You no longer have that massive delay where you have to re-engineer an application to make it compliant. The hurdles of data access, security reviews, and operational standards are addressed as the app is being built, not after.
While many cloud vendors offer AI code generation, it’s often a standalone tool. Domo’s approach seems to be about providing an integrated stack with the database and governance layer included. What are the key advantages of this end-to-end flow, and can you share an example of how it accelerates a project from experimentation to a compliant business application?
That’s the key differentiator. You can get code generation from a lot of places, but most other options require you to stitch together a database, a separate code generation tool, and then a governance layer. That stitching-together process is where projects get bogged down. Having that entire stack in one place creates a seamless, end-to-end flow that is incredibly powerful. For example, imagine a marketing team wants to build an app that analyzes real-time campaign data and suggests budget reallocations. With a piecemeal approach, they’d build the logic, then fight to get it connected to the right database, and then face a lengthy review to ensure it respects customer data privacy. With this integrated flow, they can use a natural language prompt to start building the app, it’s already connected to their curated campaign data in Domo, and the outputs are automatically governed by the existing security rules. This integration is what allows them to move from a rough idea to a legitimate, compliant business application in a fraction of the time.
Beyond just simplifying the initial build, there’s a growing need for applications that can execute tasks without constant human intervention. What do you see as the potential for a tool like App Catalyst to evolve from being an interactive assistant to becoming a truly autonomous agent that can trigger its own workflows?
That is the natural and most exciting evolution. Right now, we’re simplifying the build process, which is a massive step forward. But the true game-changer will be when these applications become smart enough to act on their own. The foundation is already there. Because these apps are built on a platform with real-time data, they can be designed to monitor for specific changes or thresholds. The next step is to empower them to not just alert a human, but to trigger their own workflows based on those data changes. Imagine an inventory management app that doesn’t just flag low stock levels but automatically initiates a reorder process with a supplier based on sales forecasts. That’s the future: moving from interactive tools that assist humans to autonomous agents that execute business processes, freeing up people to focus on higher-level strategy.
Your roadmap mentions expanding something called the Model Context Protocol to connect agents developed in Domo with other AI platforms. Can you break down how this works and maybe give an example of how it would enable more complex, cross-platform AI processes for your customers?
The reality of the enterprise is that no single platform does everything. Our customers use a variety of AI tools and platforms, and they need them to work together. The Model Context Protocol is our framework for enabling that communication. Essentially, it acts as a universal translator, allowing an AI agent built in Domo to securely share context and data with an agent built on another platform, and vice versa. For instance, a customer could have a Domo agent that monitors supply chain disruptions. When it detects a delay, it could use the protocol to trigger an agent in their ERP system to adjust production schedules, which in turn could signal an agent in their CRM to proactively notify affected customers. It enables a more sophisticated, orchestrated business process that spans multiple systems, creating a truly intelligent and responsive organization.
What is your forecast for the business intelligence and AI platform market?
I believe we’re going to see a major consolidation of capabilities. For years, the market was fragmented—you had tools for data visualization, separate tools for data prep, and AI was this niche thing off to the side. Now, generative AI is forcing a convergence. Platforms that can offer an integrated, end-to-end flow from raw data to a governed, AI-powered application will pull away from the pack. The future isn’t about having the best dashboarding tool; it’s about providing a unified environment where businesses can create, deploy, and manage intelligent applications securely and at scale. We’ll also see a huge emphasis on planning and simulation. It’s not enough to analyze what happened; businesses need to evaluate alternative scenarios and potential outcomes. The vendors who can combine analytics with robust planning capabilities will be the ones that truly empower the next generation of data-driven enterprises.
