Skip to content
Since98
Insights · AI & Automation

How to build an AI agent for business: seven steps from task to monitoring

To build an AI agent for business, start with one repetitive task and collect real examples of it. Give the agent the narrowest access that does the job and test it against those examples. Then decide which steps a person approves and keep logs after launch. The model is the easy part; data, permissions and testing decide whether it works.

02 — Short answers

Short answer

How do you build an AI agent for a business?

Short answer
Pick one task that repeats and has clear inputs, and gather real examples of it. Connect only the systems the task needs and write the rules and limits. Test on your own cases, add approval steps for risky actions and monitor the logs after launch.
How we do it
This is the sequence we follow on agent projects; each step ends with something written down that the business owner signs off.
Limits
If the task changes every week or nobody can describe what a correct result looks like, an agent will not be reliable yet. Map and stabilize the process first.

Step 1. Choose one task, not a department

Start with a single task that repeats often, has a clear input and a result someone can check. "Read incoming supplier invoices and prepare entries for the accountant" is a task. "Automate finance" is not.

Write down who does the task today, how long it takes, which systems they open and what they type. This description becomes the specification for the agent and the yardstick for testing.

Step 2. Collect real examples

Gather a set of real cases from your own work: emails, PDFs, chat messages, records, together with the correct result for each. Include the awkward ones: scanned documents, mixed Vietnamese and English, missing fields.

These examples do two jobs. They show the developer what the agent must handle, and later they become the test set. Remove or mask personal data you do not need for the task before sharing it.

Step 3. Map systems and permissions

List every system the task touches and what the agent must do in each: read, create, update or send. Then give it the narrowest access that covers that list.

Check how each system can be reached. Some local accounting, POS and CRM tools offer an API, others only file exports or nothing. Whether a given system has a usable API is confirmed on the audit, system by system.

  • Read-only where possible
  • A separate account for the agent, never a person's login
  • No permission to delete records
  • Spending, sending documents to customers and contract changes always need a person's approval

Step 4. Design the steps and the limits

Break the task into steps the agent performs, and decide for each one whether the agent acts alone, drafts for approval or stops and asks a person. Write explicit rules for what the agent must never do.

Define what happens when the agent is unsure: if a document is unreadable or two records conflict, it should hand the case to a person with a short note, not guess. This rule matters more than the choice of model.

Step 5. Test on your own cases

Run the agent on the examples from step 2 before it touches live data. Compare its output with the correct results and agree in advance what counts as a pass for each type of case.

Test in a copy of your systems or in draft mode, where the agent's actions are prepared but not executed. Repeat the tests after every change of prompts, rules or model; a change that fixes one case can break another.

Step 6. Plan the hand-off to people

Decide who receives the cases the agent cannot handle, in which tool, and how fast they need to react. For a support agent this may be a staff member on Zalo; for a document agent, the accountant's task list.

Tell the staff what changes for them. People who know which cases the agent sends them, and why, catch errors earlier.

Step 7. Launch narrow and monitor

Start with a part of the volume, one branch or one type of document, and widen it when the logs show stable results. Keep a log of every action: input, decision, output, and who approved it.

Review the logs regularly in the first weeks and then on a schedule. Models, your prices and your processes change; an agent that worked at launch needs someone who owns it and can switch it off in one step.

Personal data in prompts, logs and model provider requests falls under Vietnamese data protection rules; where data may be stored and processed is something to confirm with your lawyer.

03 — FAQ

Questions readers ask

Do I need my own AI model to build an agent?

Usually no. Business agents usually use a model from a provider through an API; the work is in connecting your systems, the rules and the tests. Which provider is acceptable for your data is part of the design.

Can I build an AI agent without coding?

For simple flows, no-code tools can work. Once the agent writes into accounting, CRM or order systems, you need someone who can read logs, handle errors and control permissions, which usually means code.

How long does it take to build an AI agent?

It depends on the number of systems and how ready the examples are. We give a timeline after the audit, which takes 2 weeks and ends with a fixed quote.

What is the most common reason agent projects fail?

The most common cause is that the task was never written down clearly, so nobody could say what a correct result is. Steps 1 and 2 exist to prevent that.

04 — Sources

Sources and notes

  1. 01Our audit durationOur published prices and terms on this site
  2. 02Agent project stepsStudio practice for agent projects; see the AI agents service page
05 — Author

Who wrote this

Dmitry DanilovArchitecture · AI · In web since 1998

One of the two founders of Since98. We write about what we set up for clients: what works, what it costs and where the limits are.

About the founders
06 — Articles

Read next

  • AI & Automation · 3 min read

    AI agent vs chatbot

    A chatbot answers in a conversation; an agent takes actions in your systems. How to tell which one your task needs.

  • AI & Automation · 4 min read

    Hiring an AI agent developer

    Freelancer, agency, SaaS vendor or in-house team: the trade-offs, and the questions to ask before you sign.

  • AI & Automation · 3 min read

    Agent software vs custom

    When a subscription AI agent is enough, when a custom agent pays off, and how to test the choice before you commit.

08

Next step

Describe the situation in a few lines. One of the founders replies and says whether we can help and what the first step would be.

Book an AI audit call
Or write in a messenger: