How it works, who it's for, and what to expect.

At Axecute, we run hands-on 1:1 training for people who want to use CLI agents like Claude Code, Kimi, and Codex in real work. We teach you how to give the tool enough context, guide the work, and review the result.

Most online courses teach you what CLI agents are. We care about what they can do for your actual work.

In a 1:1 session, we work through a real challenge from your organisation, building an agentic workflow to streamline a process, setting up a CRM, shipping an actual website. You leave with something usable, not a tutorial project. One of our clients came in planning to hire a vendor to build their organisation's CRM. By the end of the session, they had launched a working CRM themselves, with the skills to keep refining it.

The best way to learn is to do. Our instructors give you clear, step-by-step direction, explain the reasoning behind each move, and make sure the key concepts land through practice, not slides.

No. We run a training program. You are not buying another app from us. You are learning how to use CLI agents with more control.

No. Coding knowledge helps, but it is not required. The training is built around your current level and your actual work. The goal is to help you think more clearly, give better instructions, and use AI without guessing your way through every step.

A CLI agent is an AI tool that works through the command line, more commonly known as the terminal: a text window where you type instructions for your computer. Unlike a normal chatbot, it can read files, make changes, check the result, and explain what happened.

Chat is useful for advice. The command line is useful for action. CLI agents can work inside your project or folder, use the files you give them, and help produce something you can use.

The main tools are Claude Code, Kimi, and Codex. The exact mix depends on what you want to learn and what tools make sense for your work.

When you learn Claude Code through the CLI, you are really learning how to use CLI agents in general. Besides Claude, there are alternatives like Codex, Kimi Code, and Open Code, which let you run most popular open-source models like Qwen and DeepSeek, as well as frontier models like Grok.

Yes, it can happen. Used without care, CLI agents burn through tokens quickly. Long context, repetitive prompts, and unfocused sessions are the usual culprits.

Unless you are on the $100 plan, we would not advise relying on Claude Code alone. Our advice is to leverage a normal chat LLM like ChatGPT to help with the pre-work, then use Claude Code for the execution. Part of the training is learning how to split the work this way and scope requests so each session goes further.

We start with your goal, your current comfort level, and the kind of work you want help with. Then we walk through a real task together: giving context, planning the request, letting the agent work, reviewing the result, and repeating the process until the method is clear.

Yes. That is preferred. The training is strongest when we use real documents, projects, notes, reports, websites, workflows, or internal tasks instead of generic examples.

Yes. Feel free to reach out to your instructor with any questions after the session. We are happy to help. For more involved requests that need significant time or hands-on work, we may suggest booking a follow-up session.

Think of a CLI agent like a genie living at the root of your computer. You make a request, and it uses code to make it happen. That can mean converting files, drafting reports, organizing notes, building websites, creating internal tools, analyzing documents, cleaning up messy information, and automating repetitive tasks. Coding is one use case, not the whole point.

You should leave knowing how to frame better requests, provide useful context, guide the agent step by step, review what it produced, and identify where these tools can save time in your own work.

Yes. We train individuals, founders, managers, or small teams. We recommend keeping hands-on groups small, ideally three people or fewer, so everyone gets enough time to practise. For teams, the focus is usually shared habits: how to brief the agent, what to check, what not to trust blindly, and how to use the tools consistently.

No, we do not train large groups. Our sessions are designed for individuals and small groups of up to three people. While we do not train large groups, a general 101 or overview can be arranged for large group settings.

It is a good fit if you are curious about these tools but have not yet seen how they apply to your actual work. It is also a good fit if you tried these tools once, got weak results, and want to learn how to use them with better direction.

Fill in the contact form and tell us what you want to work on. We will follow up to confirm a time.

• Individual or small group training: $100 per pax per session (1:1 is currently available for physical sessions only)
• Large group demos or introductions: $250, online or physical

• Individual or small group training, up to 3 pax: 3-4 hours
• Large group demos or 101 sessions: 1-2 hours