Four methods for applying AI in your work

One of the core topics in our Mastering and Managing Documentation with AI course is that there are four distinct methods for applying AI in a professional context. Each one suits different team sizes, technical capabilities, and ambitions.

In this post, we outline all four and use a common scenario, of estimating a documentation project, to show how each approach works in practice. We also weigh up the pros and cons so you can decide which is the right fit for your situation.

At a glance

1. Web-based chatbot agents

A chatbot agent is like a general-purpose AI chat tool, such as ChatGPT or Claude. It’s different in that it is built to master one thing. You give it a focused purpose and connect it to a relevant knowledge base (such as documents or spreadsheets), which it draws on throughout your conversation.

Estimating example

You create an agent that uses your standard estimating spreadsheet as its knowledge base.

Ask it to calculate the time needed for each phase of a project, refining the estimate through a back-and-forth conversation.

Good for

Teams that want to iterate quickly with AI and prefer a conversational, low-pressure way to get started.

Organisations that only provide chatbot agents for staff to use.

Limitations

Output is text-based, which can feel underwhelming to stakeholders used to polished dashboards or reports.

Less impressive for senior management buy-in.

2. No-code AI app builders

These tools let you build interactive applications using plain language instructions; no coding skills required. The resulting apps are built in HTML, CSS, and JavaScript, and can be shared with colleagues or embedded in internal tools.

Estimating example

You build an interactive estimating tool where users adjust sliders for factors like document scope, content complexity, and review cycles.

The app calculates and displays the projected timeline in real time.

Good for

Teams that want something visually polished and easy for others to use.

Particularly effective for demonstrating AI’s value to senior management or non-technical stakeholders.

Limitations

Because you’re building a tool rather than an agent, the automation is assistive rather than autonomous; humans remain central to the process.

Depending on your organisation, a standalone app may also need to go through cybersecurity review and change management.

3. Command line AI systems

These tools run directly in the terminal and are designed primarily for software development workflows. They can read files, inspect repositories, run shell commands, and carry out multiple tasks in parallel. Teams with technical resources can adapt them for documentation workflows too.

Estimating example

You instruct a command line AI agent to:

1. Scan the codebase, API schemas, CLI commands, and recent pull requests to generate a “documentation scope map” listing all user-facing features that need coverage.
2. Break that scope into discrete work units.
3. Estimate complexity for each unit based on signals in the repository, such as the number of user-facing commands or screens.
4. Generate a consolidated estimate, saved as a reusable workflow for future projects.

Good for

Developer-led teams that already work at the command line and want AI to handle multiple tasks simultaneously.

Also, a strong choice for building a Python-based estimating tool.

Limitations

Requires technical confidence and is harder to share with non-technical stakeholders.

Output is text-based, with no visual interface.

4. Workflow automations

Workflow automation platforms connect different apps and services so that tasks happen automatically, often behind the scenes.

At their most powerful, they involve AI agents that can make decisions, generate content, and pass work between systems without human intervention at each step.

Estimating example

An automated estimating workflow might look like this:

1. The Project Manager publishes an Epic for a new project.
2. An AI agent identifies the documentation tasks required.
3. The agent estimates the time needed for each task.
4. It sends a summary email to the project and documentation managers.
5. The Documentation Manager reviews the estimates and assigns tasks to writers.

You could extend this further: the agent assigns work directly, drafts initial content, or flags when an Epic is missing the structured information the workflow needs to function.

In this model, expertise is captured upstream in product engineering, in the Epic itself, rather than downstream in the documentation team. The writer’s role shifts from drafting to designing and governing the knowledge pipeline. Documentation becomes, in effect, a by-product of system state.

Good for

Organisations ready to treat AI adoption as a process improvement or change management project, and willing to invest time upfront for significant returns downstream.

Limitations

The workflow depends on others (particularly engineers and product managers) including the right (structured) information in their Epics. If that behaviour doesn’t change, the automation breaks down.

Automated workflows have also historically been brittle when they encounter unexpected situations, although modern AI agents are increasingly able to handle exceptions or flag them for human review.

So which method is best?

There is no single right answer. It depends on your team’s goals, technical confidence, and appetite for change.

If you want quick wins

Web-based chatbot agents and no-code AI app builders are the lowest-risk entry points. They let your team build confidence with AI, demonstrate early results, and develop a clearer picture of where more ambitious automation could follow.

For transformational impact

Workflow automations offer the most significant potential returns, but they require more planning, stakeholder alignment, and change management.

Treat implementation as you would any major process improvement project: identify the biggest problems, assess whether automation can address them, and pilot before you scale.

Consider command line tools if

Your team has technical resources and wants AI to work across multiple tasks simultaneously, particularly where integration with a codebase or repository is valuable.

If you need help

Cherryleaf can help you choose and implement the right approach through our training courses and AI consultancy services. Contact us to discuss your situation.