Understand when mentoring can help
Mentoring can help when you have a concrete decision, a project in development or a difficulty you cannot resolve alone. Bring enough context to discuss alternatives and consequences. Support becomes useful when the conversation ends with a decision you can try and later review against what actually happened while applying it to your work.
You may be building an application, organising an automation or trying to understand why a service requires too much work for the margin it leaves. Those questions connect technology and business. A tool may solve part of the execution, but the choice depends on the offer, people and conditions in which you will work.
In my 2022 book, I described lessons about demand for support when applying knowledge. That principle remains part of how I think about training. Access to explanations helps; bringing a concrete attempt makes it possible to discuss where application became difficult and choose what is worth changing next.
If you want someone to take responsibility for implementation, start with AI consulting for businesses. In mentoring, you bring your work for discussion and remain responsible for developing it. That distinction should be clear before choosing support, so you understand what to prepare and what to expect from the format.
What I work on in Laboratório da IA mentoring
At Laboratório da IA, I lead two monthly mentoring sessions about online business and artificial intelligence, reviewing students’ projects. Questions may concern strategy, processes, tools or execution. Each project’s context guides the conversation, so suggestions address the need presented and can later be evaluated through practical work carried out by the participant.
Someone may bring an application that already works and need to review permissions or the user journey. Another participant may want to automate a task without having decided how to organise the data. A question about the offer, acquisition or delivery capacity may also be limiting the project’s progress.
In Sextas Ímpares #137, dedicated to Laboratório da IA with guests, you can hear a discussion about applying tools and participants’ experiences. The Portuguese-language broadcast documents that moment. For the current offer, this page refers to the two monthly sessions I lead and the conditions available on the Laboratory website.
You can explore my teaching approach through classes on building applications with Claude Code and advertising-management decisions in the AI era. They illustrate reasoning that connects execution to business choices, without promising that one session resolves every problem in a project. Those classes are also in Portuguese.
Prepare a question that allows work on the project
A useful mentoring question presents the objective, current situation, previous attempts and decision you need to make. Include examples prepared for sharing and identify relevant constraints. That preparation lets the discussion focus on examining the work, rather than reconstructing the entire history from the beginning before reaching the difficulty you want to address.
Instead of only asking which AI tool to use, explain the task. What information comes in? Who needs the result? Where is it stored? Which error would have consequences? If you have already tried a solution, show an example that worked and another that failed, removing private data and credentials.
For a sales question, bring the offer, audience and stage where you experience difficulty. Perhaps too few enquiries arrive, people do not understand the proposal or delivery takes too much time. The question becomes more concrete when you distinguish those situations and show what you observed in actual conversations.
The pages on process automation and AI data analysis help prepare examples. You do not need to present a perfect project. We need enough information to understand the difficulty and choose an experiment that can teach something useful about it, including whether the first explanation was incomplete.
Leave with an experiment you can carry out
After mentoring, choose a change you can execute and observe before adding further workstreams. Define the expected result, necessary precautions and information you will keep. That experiment makes it possible to return with evidence, including what failed or introduced an unexpected difficulty that could not be understood from the initial discussion alone.
Imagine a project struggling to route enquiries. The next step may be reviewing three examples and writing clearer criteria before rebuilding the whole automation. If the difficulty concerns the proposal, testing a different explanation in a conversation and recording the questions it raises may be a more useful first action.
In Sextas Ímpares #147, during the discussion of investing in learning, I return to the difficulty of accumulating training without applying it. That concern shapes what I expect from support. I want work between conversations, so we can discuss observed changes rather than simply adding more possibilities to the list.
Keep the reason for choosing the experiment as well. If the outcome differs from expectations, that note helps identify which assumption failed. The article on growth, margins, teams and AI develops the connection between learning, applying and assessing business consequences, which gives the next conversation more substance.
Connect technical decisions with the offer and operations
A technical decision should be assessed through the work it enables and the responsibilities it creates in the business. Consider who uses the solution, who maintains it and how it contributes to delivery. Mentoring connects the tool with the objective, so construction stays aligned with what you want to sell or improve in practice.
If you are developing an application to support a service, examine whether customers need that journey and whether the team can maintain it. A new feature may impress in a demonstration while changing little about delivery. Another, less conspicuous change may resolve a difficulty that appears in every project.
The page on developing applications with AI covers planning, the first version and maintenance. The article on planning with Claude Code develops preparation. These readings help you bring more specific questions about users, data and expected behaviour when that is the subject of the project you want to discuss.
There are also decisions about stopping or simplifying. Time already invested in a direction deserves to be understood, but the next decision should consider what remains worth building. That business principle, present across the book and classes, supports clearer discussion of priorities and future effort instead of continuing solely because work has already been done.
Choose the learning format that fits the need
Laboratório da IA provides ongoing support around the business and AI subjects I work on. Corporate training addresses a team’s needs, while a talk has a different duration and objective. Choose the format based on the work you want to develop, the people involved and the time available to apply the learning afterwards.
I have worked in digital since 2012 and have had more than 30,000 students throughout my training career. That experience includes marketing and online business alongside current work with AI. The About page presents the background and projects that help explain where my questions and evaluation criteria come from.
If you want to prepare a team around a shared objective, explore AI training for businesses. If the need concerns an event, see talks about artificial intelligence. Conditions are discussed according to context, without treating every format as the same programme or assuming identical preparation needs.
To explore ongoing support, visit Laboratório da IA. Check the current conditions and consider the project you would like to work on. A useful starting point is being able to explain what you want to improve and reserving time to try what makes sense after the discussion.