Over the last year, it has been almost impossible to ignore the rapid pace at which AI is evolving. Every week seems to bring a new model, a new framework, or another AI tool that promises to transform the way we work. Like many technology professionals, I was excited to explore these advancements and understand how they could fit into my daily work as a Microsoft Dynamics 365 Business Central consultant.
My journey began with simple experiments. I started using AI to generate documentation, review AL code, summarize technical information, explain complex concepts, and even help troubleshoot Business Central issues. I tested different AI models, compared their strengths, experimented with prompts, and explored various ways to improve the quality of the responses. Each experiment taught me something new, and with every success, I became more convinced that AI could become an invaluable part of our profession.
However, after several months of experimentation, I realized that I was asking the wrong question.
Initially, my focus had been on how AI could make me more productive. How could it help me complete tasks faster? How could it reduce repetitive work? How could it improve my own efficiency as a consultant and developer?
Those were useful questions, but they were too narrow. The bigger opportunity was not about making one consultant more productive. It was about transforming how an entire Business Central team works.
That realization completely changed my perspective. Instead of thinking about individual productivity, I started thinking about how AI could become part of our delivery process. Rather than building isolated AI tools for myself, I began exploring how specialized AI agents could support consultants, developers, testers, and support engineers throughout the lifecycle of a Business Central project.
Today, my focus is no longer just on building AI agents. My goal is to build an AI-first Business Central practice, where AI is not treated as another application sitting on someone’s desktop but as a trusted assistant that naturally becomes part of the way the entire team collaborates, solves problems, shares knowledge, and delivers value to customers.
I believe this is where the real transformation begins—not when one person becomes more productive, but when an entire team learns how to work alongside AI in a structured, consistent, and meaningful way.
AI Is More Than Just Another Tool
As I started building more AI agents, I noticed something interesting within my own team. Like many organizations, we had access to powerful AI tools such as GitHub Copilot, ChatGPT, Claude, and Microsoft Copilot. Everyone was excited to experiment with them, and initially, I encouraged the team to explore as much as possible.
The results were mixed. Some team members quickly discovered how AI could help them write better code, generate documentation, or solve technical problems much faster. Others found it difficult to understand where AI could genuinely help in their daily work. Everyone had their own prompts, their own workflows, and their own way of interacting with AI. While there were individual success stories, I realized we were missing something important. Knowledge was staying with individuals instead of becoming part of the team’s collective capability.
If one developer discovered a great prompt for reviewing AL code, another developer might never benefit from it. If someone figured out a better way to generate documentation, that learning often remained on their laptop instead of becoming part of our standard way of working. That made me rethink our approach. I didn’t want AI to become another personal productivity tool that everyone used differently. I wanted AI to become part of the way we work together.
The question was no longer, “How can each person use AI?” Instead, it became, “How can we build a consistent AI-driven way of delivering Business Central projects?”
Moving Beyond Prompts to Agent Design
When I first started exploring AI, like many developers, I spent a lot of time experimenting with prompts. Every discussion around AI seemed to focus on prompt engineering. The better the prompt, the better the result and to be fair, prompts do matter but after building multiple AI agents for Business Central, I realized that prompts are only a small piece of the overall solution.
The real challenge is not writing clever prompts. The real challenge is designing an agent that consistently solves a business problem. Today, when we discuss AI within the team, we rarely begin by asking, “What prompt should we write?” Instead, we begin by asking much bigger questions.
What business problem are we trying to solve?
Who will use this agent?
What role should it play within the project?
What Business Central knowledge does it need to understand?
Which tools should it have access to?
How should it investigate a problem before providing an answer?
When should it stop and ask for more information instead of making assumptions?
What should success look like for the user?
For me, this has become one of the biggest mindsets shifts in our AI journey. The goal is not to create people who know how to write better prompts. The goal is to build a team that understands how to design AI agents that solve real Business Central problems in a consistent, reliable, and scalable way.
Building Specialized AI Agents and Shared Knowledge
As I continued building AI solutions, I quickly realized that one AI assistant could not effectively solve every Business Central challenge. Documentation requires a different approach than troubleshooting. Reviewing AL code demands different knowledge than generating UAT scenarios or analyzing business processes. Trying to combine everything into a single assistant only made the solution more complex and less reliable.
Instead, I started building specialized AI agents, each with a clearly defined role, its own instructions, workflow, and tools. Every agent is designed to solve one problem well rather than trying to solve every problem reasonably but building specialized agents was only part of the journey. I also wanted every improvement to benefit the entire team.
Whenever we refine an instruction, improve an agent’s workflow, or discover a better way to solve a Business Central problem, that knowledge becomes part of our shared AI ecosystem instead of remaining with one individual. Over time, we are not just building better AI agents—we are building a reusable knowledge base that grows with every project, making the entire team more consistent, more efficient, and better prepared for future challenges.
AI as a Digital Team Member
Perhaps the biggest mindset shift for me has been learning to stop thinking of AI as just another tool. Instead, I have started thinking of AI as another member of the project team.
Every Business Central implementation already relies on specialists. We have functional consultants who understand business processes, developers who build solutions, testers who validate functionality, project managers who coordinate delivery, and support consultants who resolve issues after go-live. I believe AI agents will gradually become specialists alongside them—not replacing people, but supporting them throughout the project lifecycle.
Imagine a project where documentation doesn’t begin after development is complete because the Documentation Agent has already prepared the first draft. Before a developer requests a peer review, the Code Review Agent has already identified potential issues and improvement opportunities. When a customer raises a support ticket, the Troubleshooting Agent has already gathered relevant information, highlighted possible root causes, and suggested where the investigation should begin. Even before UAT starts, a Testing Agent has prepared structured test scenarios based on the implemented functionality. The role of the consultant doesn’t disappear. It evolves.
Instead of spending hours gathering information, formatting documents, reviewing repetitive code patterns, or preparing test cases, consultants can focus on what they do best—understanding business requirements, solving complex problems, making informed decisions, and guiding customers. This is the vision I am working towards. Not a future where AI replaces consultants, but one where every consultant has a team of intelligent digital assistants working alongside them. Just as we collaborate with colleagues who have different areas of expertise, I believe we will soon collaborate with specialized AI agents that understand documentation, development, testing, troubleshooting, and business processes.
When repetitive work is delegated to AI, people gain more time for creativity, innovation, customer engagement, and strategic thinking. For me, that is what an AI first Business Central practice truly means—not simply using AI to complete tasks faster but building a workplace where people and AI collaborate seamlessly to deliver better outcomes for every customer.
Building an AI First Mindset
Throughout this journey, one lesson has become very clear to me: technology alone will never build an AI first practice people will. That is why my focus is not only on creating better AI agents but also on helping my team embrace a new way of working. I encourage curiosity, experimentation, and continuous learning, while ensuring that every new discovery becomes part of our shared knowledge through better instructions, reusable workflows, and specialized AI agents.
My goal is not to create a few AI experts within the team. It is to build a culture where AI knowledge is continuously shared, refined, and improved so that everyone benefits from what we learn together. I believe the future of Business Central consulting will not be defined by who has access to the latest AI model. It will be defined by the teams that learn how to integrate AI into every stage of project delivery—from requirements gathering and development to testing, documentation, support, and customer collaboration. This is the journey I have started with my own team. We are still learning, still experimenting, and still improving. Every new AI agent, every project, and every conversation with the community helps us move one step closer to that vision.
My ambition is not simply to build AI tools. It is to build an AI first Business Central practice where consultants, developers, and intelligent agents work together as one team—reducing repetitive work, accelerating delivery, and allowing people to focus on what truly creates value: solving business problems and helping customers succeed because I believe the future of Business Central consulting is not just about adopting smarter technology.
It is about building smarter teams that know how to collaborate with AI.
Leave a Reply