Augmented Expertise is a modern methodology for knowledge workers. It uses AI to document and amplify the professional’s existing expertise, rather than replacing it.
The origin story: A strong desire to document my knowledge
For most of my career, I’ve felt a strong desire to document my knowledge and experience.
My immature vision was a wiki-style knowledge base with articles about all topics that I’d built real expertise in over the years.
For obvious reasons, this never materialised: Writing down my entire expertise would have taken just as long as acquiring it.
AI did not feel like a natural extension of me and my work
With the rise of AI, I didn’t see the connection at first. I tested AI tools early on for automation, vibe coding, writing, and other experiments.
None of it felt like a natural extension of me and my work.
The magic moment: Skills
When I first used skills (reusable instructions for AI) in early 2026, it finally clicked. This was the unlock I had been waiting for, without knowing it.
Skills allow me to easily document my knowledge, my workflows and my approaches to problem solving in a format that AI can use.
Why my wiki vision was wrong and AE is better
The biggest advantage of Augmented Expertise is that the act of documenting and creating instructions itself does not require much additional effort.
It happens naturally while you work, and at the end of the day, it helps you save time while generating compounding value, instead of taking up more of your time.
This is what makes Augmented Expertise so much more powerful than writing a wiki-style knowledge base: With AE, the wiki writes itself.
How does Augmented Expertise really work?
Augmented Expertise does not require an additional time investment, but it does require a new approach to work.
Now, you embrace AI, and you use it like a very intelligent but inexperienced and overly eager colleague that you’re training.
You explain your work, your workflows, and your reasoning, in order to create outputs and deliverables with AI that the pre-AE version of you would be immensely proud of.
What do you get from working with Augmented Expertise?
The result of every AI work session with Augmented Expertise is the deliverable or output you worked on plus a skill or several skills that capture everything the session surfaced.
When you work on something similar next time, you do not have to explain the same things to your AI again. You just pick it up a few steps ahead of where you left it and your documented knowledge and workflows compound over time.
The compounding value of Augmented Expertise
In following sessions, when skills already exist for the work that you’re doing, the skills do not stay stale.
Based on your corrections to the AI’s output, your further explanations, and edge cases you encounter while doing the same type of work on new cases, your skills are automatically improved and refined.
Augmented Expertise is more than just skills
Skills are currently the ideal format for documenting and formalising a knowledge worker’s expertise, but they are not the only way:
Depending on which AI you use, there are workspaces, projects, knowledge bases and other concepts.
Skills might not even survive, but the approach will last: Observing your work and documenting your expertise in the background to make it reusable.
Why you should start using Augmented Expertise today
I’ve documented and open-sourced the entire framework as a meta-skill. The name “meta-skill” means that the approach itself lives in a skill, and it creates and improves skills as the format that your expertise is documented in.
In the first three months of using Augmented Expertise, it helped me create more than 40 skills for different areas of my work and it applied more than 600 skill improvements that were captured during my real work sessions.
I also noticed an immediate increase in my income: Augmented Expertise in practice meant more billable deliverables per hour worked, without a ramp-up period.
How to get started with Augmented Expertise
Are you still unclear about which AI to use for this approach? My personal recommendation would be Claude Cowork. That’s where I’ve built and refined this methodology since the beginning of 2026.
If you’re already using an AI that is not Claude Cowork, you can still start with Augmented Expertise today: If your system supports skills, the methodology is fully portable. And even without skills, it can be adapted to whatever knowledge base format you want to use.
Whatever your setup, the act of getting started is itself your first task with the Augmented Expertise mindset: Take the content of the GitHub repository, feed it to your AI and tell it that from now on, this is your approach to work. Your AI is able to guide you towards the ideal implementation for your specific setup.
Your first steps with Augmented Expertise
The GitHub repo contains a detailed user guide with guidance on how to get started and be successful with the meta-skill. I encourage all users to ask questions and provide feedback and I’m more than happy to help with issues that might come up.
And in case you’re still not convinced: The most beautiful thing about the meta-skill is that it also improves itself over time. This is self-improving Augmented Expertise, or meta-AE, if you insist.
The principles of Augmented Expertise
- AI will not take over your thinking. With Augmented Expertise, you will be thinking more, harder, and over longer periods of time. You’re a knowledge worker, so thinking is your strongest tool. AE gives you the freedom to use it.
- You get the most value out of AI by using it for things that you were already excellent at without AI. Augmented Expertise only works when there is expertise to augment.
- You own your output and deliverables. If the AI makes errors, they are your errors. Treat everything you produce as something you are willing to defend.
- Transparency is key. There’s no point in hiding that you use AI. Explain how you use it and explain Augmented Expertise. It’s your strongest asset, not something to be ashamed of.
- AI will not replace knowledge workers. It will make our work more fulfilling by freeing up time for the things we enjoy most and amplifying the impact of our expertise.
The next steps for Augmented Expertise
This approach has transformed the way I work and I’m excited about sharing it with other knowledge workers and seeing how you adapt it. Please reach out to let me know!
I will also implement Augmented Expertise at the team level in several organisations over the next few months, and I’m curious to see how it compounds beyond individual professionals.
Sign up here if you want to be notified when there is important news to share about Augmented Expertise: