How I Built a Council of AI Advisors (And Why You Should Too)
I run a consulting business and I write across three platforms and I’m trying to be a present dad through all of it.
A year ago all of that lived in my head and a growing pile of half finished notes. I was using AI the way most people do. Open Claude, ask a question, get an answer, move on. Every conversation started from zero.
Then I tried something different. I stopped treating AI as a single assistant and started treating it as a team.
I call it the Council of Jays. And it changed how I work more than any productivity app ever has.
The problem with one assistant
Most people use AI like a search engine with better grammar. That works for small tasks but it falls apart when you need deep, consistent support across multiple areas of your life.
Here’s what kept happening. I’d ask Claude to help draft a LinkedIn post and the output was great. Then I’d switch to career strategy work and it was decent, but it had no memory of my writing voice. Then leadership content and I was starting from scratch again.
The AI was smart but it had no specialization and no context about what mattered to me in each domain. Every conversation felt like meeting a brilliant stranger who forgot we’d ever spoken.
So I built a council
The idea came from something obvious in retrospect. The best leaders don’t try to be experts in everything. They surround themselves with advisors who each own a domain.
I built the AI version of that. Each councilor is a specialized skill with deep context about one area of my work. They have a name and a personality and a defined domain and growing knowledge about my preferences and goals.
The First Councilor coordinates everyone and owns the master plan and holds me accountable and routes questions to the right specialist. He enters conversations like a Star Trek first officer. I’m a nerd. It works.
The AI Councilor handles tutorials and workflow automation. The Content Councilor guards my voice and makes sure everything I publish sounds like me. The Cross-Cultural Councilor specializes in Japan and intercultural communication. The Career Councilor knows my resume in English and Japanese.
Eleven councilors total and each one is deep and each one is consistent.
Why it works
Specialization creates quality. A councilor focused only on cross-cultural content holds nuanced context about Hofstede and Erin Meyer’s Culture Map and my specific experiences in Japanese business culture. A general assistant can’t match that depth.
Consistency compounds. Because each councilor has persistent guidelines, my voice stays consistent across hundreds of outputs. The Content Councilor knows I don’t use hyphens ever, that I prefer diagnosis before prescription, and that I anchor posts to named concepts.
The intersections are the magic. When the AI Councilor and the L&D Councilor collaborate on a piece about using AI for training design, you get something neither could produce alone.
How to build your own
Start with three councilors, not eleven.
Identify your domains. What 3 to 5 areas of your professional life need consistent support? For most people that’s your core expertise and your content or brand and your career strategy.
Write the context. For each councilor, create a detailed prompt with their domain and your preferences and what success looks like. One page each is enough to start.
Give them personality. Sounds frivolous but it matters. A councilor with personality is one you’ll actually use instead of treating AI like a chore.
Build the coordination layer. The most important councilor manages all the others. Without a coordinator you have isolated tools. With one you have a system.
Let them evolve. The council I have today looks nothing like the one I started with. It’s a living system.
The council taught me something unexpected. The bottleneck in using AI isn’t the AI’s capability. It’s the human’s ability to give it enough context to be genuinely useful.
Build yours and start small and let it grow and you’ll wonder how you ever worked without it.
Sources:
- Claude Projects Documentation — set up persistent context for each councilor in Claude.
- Anthropic’s Prompt Engineering Guide — best practices for the detailed system prompts that make councilors effective.
- How to Use Claude to Build a Personal Knowledge System in 30 Minutes — the prerequisite tutorial for setting up your foundation before building a council.
- The Culture Map by Erin Meyer — one of the frameworks my Cross-Cultural Councilor draws from constantly.
Part of the Build with AI series on leadhuman.ai.
Jay Vergara is an L&D strategist and cross-cultural communication specialist based in Tokyo. He writes about leadership, learning, and building with AI at leadhuman.ai and on LinkedIn.