I spent months saving AI guides instead of starting

For months, I saved YouTube videos, LinkedIn frameworks and Substack guides. I kept thinking: “One day I will read it all and start.”

That day did not happen for months.

Finally, I made a decision: “I will buy a laptop, download Claude and start.” I was tired of feeling behind.

I took small steps each day. I used templates and YouTube videos, then I started building. Slowly, I saw that it was possible and that I could do it.

The biggest breakthrough came when I understood how AI could help in my daily life, learning and work. I have not stopped since.

I am writing this for the experienced employee who reads about AI but still keeps the subject at a distance. Maybe you have spent 20+ years in the same company. You know the customers, the market and the history behind decisions that look simple from outside.

I can see several ways an employer may use the capacity that AI creates. An overloaded team may return to normal hours. A team with manageable work may receive a broader scope, including projects it had delayed because there was no capacity. The simple, manual parts of the work may become lighter, leaving people with more difficult decisions. A planned hire may not happen. A role may change. And if there is no extra work, a company may decide it does not need the same number of full-time roles.

This is the scenario that worries me most for experienced employees. Headcount reduction after 20 years in one company is a hard reality. It affects income and confidence, and it makes the career you built feel less secure.

When a company introduces AI into its work, management needs to explain why it is doing it, which work will change, the skills that will matter and the point at which decisions will be made. Employees need this clarity before they can make sensible decisions about their own work.

Your own first move is to learn how this new way of working can work in your role.

Your experience is still one of the most valuable resources you have today. You have 2 or 3 decades of subject-matter expertise. You hold the judgment calls. You know what is right and wrong for your position and company. You know when a number looks correct but the conclusion is wrong. You know which customer issue needs attention now and which one can wait. You know the difference between an interesting idea and a useful commercial decision.

AI can help with the work around these decisions. It can help you search and summarise, compare big documents, flag anomalies, prepare a first structure, or repurpose something you have already created. You remain responsible for the decision.

That responsibility is the point.

Your domain knowledge and commercial judgment are already built. AI fluency is the third pillar. It gives your existing experience another way to work.

Start with your own day

Take a sheet of paper and map one normal process. What exactly do you do? How do you do it now? Where are the bottlenecks? Which steps require a lot of manual input and still do not bring substantial value, such as judgment or decision-making?

Do this privately first. You do not need to tell anyone. You do not need to create a presentation. You are learning your own work in a new way.

Then ask AI to help you redesign this one process.

You might find that some steps happen on autopilot. You might see that you search for the same information in several places, or that you prepare the same kind of summary every week before you can make the real decision.

The first value is clarity. You can see what you actually do with your time.

The second value comes when you test one small improvement. You will learn where AI is useful, where it creates more work and where your own judgment must stay close to the process.

Here is a private exercise you can copy into a note and use this week.

My private workflow audit

1. Choose one recurring piece of work from this week.
2. Write every step down as you currently do it. Include searching, reading, checking, writing, chasing information and preparing the final output.
3. Mark the steps that take time but do not require your commercial judgment.
4. Write what a good result looks like. Be specific. What must be correct? What would make the result useless? What needs your final review?
5. Give the process to your approved AI tool, using a safe and non-confidential example if required, and ask: “How could this process be redesigned so I spend more time on judgment and less time on repeated manual work?”
6. Test one small change. Keep the review with you. Write down what worked, what did not work and what you would change next time.

This exercise is small on purpose. It builds your own understanding of how AI could become part of your work.

The practical start

I built a free guide, Claude Code in six steps, for this exact first part of the journey.

It takes about 30 minutes. You set up the desktop app, explain who you are and what your work looks like, audit one real week, then make one task repeat. The final step shows how to write prompts that give the tool the context and standards your work needs.

The guide keeps the starting point concrete. You work on the job you already have, with your own real tasks and your own judgement.

Even 30 minutes a day is enough to begin. Use that time to understand a tool and one type of task. Learn the limitations as well as the possibilities, and explain what good work means in your field.

The practical skill is bigger than prompting. It is learning how to work alongside AI, manage its output and decide where it belongs in your day.

When you can do this with one real process, you will have something much stronger than a certificate from a general training course. You will have evidence from your own work.

You may keep this private. That is a good first stage, especially when you are still learning and want space to make mistakes.

Later, you can choose to show the right people what you have learned. A person who can identify a bottleneck, test a safe AI-supported process, explain the business value and stay accountable for quality becomes visible in a different way.

This was my own path. I invested my own time and resources after work to learn and experiment. I could show that I was ready for an AI-focused role.

I began as an experienced commercial person who was tired of feeling behind.

My earlier five-stage map starts with the blank chat. It ends with an AI-enabled leader who can design a working model for other people. Take the next move from where you are now.

The pace of change can feel uncomfortable. Your existing experience still matters. It becomes more useful when you can connect it to an AI-enabled way of working.