Like many people exploring Artificial Intelligence, I initially looked at AI as another productivity tool. My goal was simple: find ways to work faster, reduce repetitive tasks, and improve the quality of my deliverables.
I started experimenting with AI to write documentation, review AL code, generate UAT scenarios, summarize meetings, explain Business Central functionality, and even brainstorm solution designs. Every new experiment revealed another opportunity where AI could save time and help consultants focus on higher-value work. The results were genuinely impressive. Tasks that previously took hours could now be completed in minutes. AI could generate well structured documentation, review code for potential improvements, explain complex Business Central concepts, and even suggest better ways to approach business processes. Like many others, I became fascinated by what these new AI models were capable of.
For a while, I measured success with one simple question:
“Can AI do this task?”
Every successful experiment felt like another confirmation that AI would transform the way we work, but as I spent more time building AI agents specifically for Microsoft Dynamics 365 Business Central and more importantly, using them in real consulting scenarios I began to notice something. Two AI assistants could complete exactly the same task, yet one would produce recommendations I would confidently use with a customer, while the other felt generic, inconsistent, or disconnected from the realities of a Business Central implementation.
That made me pause.
Maybe capability wasn’t the real measure of success. Maybe the real challenge wasn’t whether AI could perform a task. The breakthrough came when I stopped treating AI like another software tool.
Instead, I started treating it like the newest member of my consulting team. That simple change in perspective transformed the questions I asked, the way I designed AI agents, and ultimately the way I think about AI adoption in Business Central. Instead of building tools that simply generate answers, I began designing AI team members with clear responsibilities, the right knowledge, structured workflows, and a defined role within the implementation lifecycle.Looking back, I believe that mindset shift has been the most important lesson of my AI journey so far.
Every AI Agent Needs a Job Description
Once I started thinking of AI as a new team member, the way I designed AI agents completely changed. If a new consultant joined my Business Central practice, I wouldn’t immediately assign work without first defining their role, responsibilities, required knowledge, tools, and how success would be measured. Yet when many organizations adopt AI, they simply provide access to ChatGPT or Copilot and expect people to figure out how to use it effectively.
I believe AI deserves the same level of planning as any new employee.
That’s why every AI agent I build starts with a clear job description rather than a collection of prompts. Before writing a single instruction, I define the business problem it will solve, who it will support, the Business Central knowledge it requires, the tools it should use, the decisions it can assist with, and when it should ask for more information instead of making assumptions. Only then do I design its instructions and workflows.
Just as people perform better when they clearly understand their responsibilities, AI agents become significantly more reliable when they have a well-defined purpose and role within the Business Central implementation lifecycle.
Specialized AI Agents Need the Right Context
One of the biggest lessons I learned is that successful Business Central projects are built by specialists. Functional consultants, developers, architects, testers, support consultants, and documentation specialists all bring different expertise to the project. AI should be no different. Instead of building one assistant that tries to do everything, I started designing specialized AI agents each with a clear responsibility, the right instructions, and the appropriate tools. Whether it’s a Documentation Agent, UAT Agent, Code Review Agent, or Troubleshooting Agent, focused agents consistently deliver better results than a single general-purpose assistant.
However, specialization alone isn’t enough. Just like every new Business Central consultant needs onboarding, AI agents also need context. They must understand your implementation methodology, development standards, documentation templates, customer processes, and organizational best practices. Without that context, even the most advanced AI models produce generic recommendations. With access to trusted organizational knowledge, they begin to behave like experienced consultants who understand not just Business Central, but also the way your organization delivers successful implementations. That’s why I believe every Business Central practice should invest in an AI Knowledge Hub a shared knowledge foundation that enables every AI agent to learn, improve, and deliver consistent recommendations.
AI Becomes Part of the Team
As I built more AI agents, I realized they should be managed just like any other member of the consulting team. Every agent needs a clear responsibility, defined workflows, access to the right knowledge and tools, and continuous improvement. More importantly, they need to understand their limitations. Just like experienced Business Central consultants, trustworthy AI agents shouldn’t guess—they should investigate, ask for clarification when needed, and make recommendations only when they have sufficient context.
This also changed the way I think about the future of AI in Business Central. I don’t believe AI will replace consultants; I believe it will strengthen them. Specialized AI agents will prepare documentation, review code, generate UAT scenarios, organize support investigations, and retrieve organizational knowledge before a consultant even begins working. The consultant remains responsible for every decision, while AI removes repetitive work and accelerates preparation. In that future, AI isn’t a replacement for expertise—it becomes another trusted member of the Business Central project team, enabling people to focus on delivering greater business value.
Looking Ahead
The biggest lesson from my AI journey isn’t about prompt engineering, language models, or the latest AI platform. It’s about mindset.
The moment I stopped thinking about AI as another software tool and started treating it like a member of the consulting team, everything changed. Instead of building generic assistants, I began designing specialists. Instead of focusing on prompts, I focused on responsibilities, workflows, knowledge, and trust. Instead of asking what AI could do, I started asking how it could work alongside experienced Business Central consultants. I believe this is where the future of Business Central consulting is heading.
In the years ahead, every Business Central practice will have a hybrid workforce where people and AI agents collaborate naturally. Consultants will continue to bring business expertise, customer relationships, creativity, and critical thinking, while specialized AI agents handle documentation, code reviews, testing, knowledge retrieval, solution analysis, operational insights, and countless repetitive activities behind the scenes.
The most successful organizations won’t simply have access to the latest AI models. They’ll have well-designed AI team members that understand their implementation methodology, learn from their organizational knowledge, and continuously improve with every project they deliver.
Business Central implementations won’t become less human—they’ll become more intelligent. Consultants will spend less time creating documents, searching for information, and performing repetitive validation, and more time understanding customer challenges, designing better solutions, and delivering measurable business outcomes. That’s the future I’m working toward. Not a future where AI replaces consultants.
A future where every Business Central consultant is supported by a team of trusted AI specialists, making implementations faster, knowledge richer, quality more consistent, and customer value significantly higher because the greatest competitive advantage won’t belong to the company with the smartest AI. It will belong to the company that builds the smartest collaboration between people and AI and I believe that journey is only just beginning.
Leave a Reply