# Agents

MAIA is a coordinated system of agents. One main agent takes your objective and directs the run. Sub-agents do the deep work on each row, in parallel, so the conversation never waits on the research.

Updated: 2026-09-30. Source: https://maia-analytics.com/docs/capabilities/agents

## Two kinds of agent

| Agent | Works on | Does |
| --- | --- | --- |
| [Main agent](https://maia-analytics.com/docs/capabilities/main-agent) | The whole project | Reads your objective, chooses the data, builds the layers, and answers you |
| [Sub-agents](https://maia-analytics.com/docs/capabilities/sub-agents) | One row each | Research one question about one place, hundreds at a time |

You talk to the main agent. Sub-agents are its research tool, and you see their work as cells filling in.

## How a run goes

1. You define an objective and dispatch it.
2. The main agent queries the data and builds a layer of candidates.
3. Where the data holds no answer, it defines a column that asks for one.
4. It tests the column on a few rows and corrects it if the answers are wrong.
5. Sub-agents research the rest, one row each. Each row spends credits, and a run above your row limit stops on an approval card first.
6. The main agent reads the finished values, checks the result, and answers.

Step 4 is why a badly worded question costs a few rows, not ten thousand. A column that sub-agents fill is an [enrichment](https://maia-analytics.com/docs/enrichments/overview).

![The chat with View activity expanded. MAIA's actions read, in part: I found 32 owner-user candidates. I'm narrowing to 18 strong candidates, then testing them in one batch to produce ten true matches.](https://maia-analytics.com/product-shots/docs/chat-activity@2x.png)

The main agent reports its test and its count before it runs the rest.

## What each one leaves behind

| Agent | Leaves | Where |
| --- | --- | --- |
| Main agent | Layers, columns, views, and a written answer | The map, the table, and the chat |
| Sub-agents | A value, its reasoning, and its sources, per cell | The table, under **View reasoning** |

While a run is in progress, a blank cell is not yet researched, and **Unavailable** is researched with no supported answer. Checking a result is covered on [Review](https://maia-analytics.com/docs/using-maia/review).

![The Layers menu on the map, listing ten layers the main agent created during the project, five of them visible, with a New layer option at the bottom.](https://maia-analytics.com/product-shots/docs/layers-list@2x.png)

What the main agent leaves behind: the layers it built, listed on the map.

## Limits

- **One run per project at a time.** Several projects can run at once.

## In this section

- [Main agent](https://maia-analytics.com/docs/capabilities/main-agent.md): MAIA's main agent takes a research objective and returns layers on the map and in the table, with a written answer. It answers from the data first, researches only what the data cannot answer, and checks the result before it reports.
- [Sub-agents](https://maia-analytics.com/docs/capabilities/sub-agents.md): MAIA is a coordinated system of agents. The main agent runs the conversation, and when a question needs research on every row, it hands the tasks to a sub-agent. Hundreds run at once, in the background, and none of them holds up the main agent.