Quick Start (3 minutes)

📚 Learning Flow

What can the AI Agent do?

Let’s see the result first—no setup, just a live demo.

Open the Open RAG Demo Site: https://openrag.dev.araliadata.io/
Paste this question and submit:

Is there a relationship between the average GDP growth at purchasers’ prices from 2021 to 2024 and the Gini coefficient of each state in Malaysia in 2024?
The demo will automatically run five Agents (Search → Plan → Decide Filters → Execute Query → Interpret) and return:
- a source list of the datasets used (from multiple Planets such as SDG, UN), and
- an aligned insight that compares the same years and geographies.

Behind the scenes, the LLM interprets your question and calls the Agents to search the Aralia Data Ecosystem, retrieve the right data, align it, and summarize the finding.

Try these next (sample questions)

Malaysia • GDP growth × Gini (2019 / 2021–2024)

What is the average GDP growth rate of each state in Malaysia in 2019?
Is there a relationship between the average GDP growth at purchasers’ prices from 2021 to 2024 and the Gini coefficient of each state in Malaysia in 2024?

Traffic safety • DUI & crashes

Which county/city has the highest number of alcohol-related traffic fatalities within the breath-test range 0.16–0.25 mg/L (or blood 0.031%–0.05%)? Rank by county/city.
Within the six major municipalities, how many fatalities and injuries occurred in the specified alcohol range?
Which road category accounts for the most alcohol-related fatalities?
What is the average age of drivers with alcohol level > 0.80 mg/L (or blood > 0.16%)?

Tip: If a dataset uses different geographic levels or units in your country, adjust the wording (e.g., region vs. state, BAC units) accordingly.

How to ask a

complete

question (must-read)

State these four pieces clearly so the Agent can answer without follow-ups:

Metric / Definition — what you’re measuring (e.g., average GDP growth at purchasers’ prices, Gini coefficient).
Time — a specific year or range (e.g., 2021–2024, 2024).
Geography / Scope — country, state, county, age group, etc. (e.g., Malaysia by state).
Relationship / Comparison — correlation, ranking, gap, or grouped comparison.

Checklist (copy and tick):

Metric/definition is clear
Year or time range is clear
Geography level and scope are clear
Relationship/comparison is clear

🧯 Common misconceptions

“Will the Agent guess missing years/regions?” → No. Please include them in your question.
“Will the LLM fetch data from the open web?” → No. It queries the Aralia Data Ecosystem and interprets the returned results.
“Do I need to code?” → No. The Colab notebook wraps the workflow. (Advanced users can explore the Agent SDK later.)

Curious how the Agent achieves this? Continue to the next chapter for the key concepts and system overview.

                                                ← Previous： [AI Agent Tutorial](<https://deciduous-centipede-9d7.notion.site/AI-Agent-Tutorial-264ddf94fd14808fafaee7019ca998c2>)　|    Next：[

System Overview & Core Concepts](https://deciduous-centipede-9d7.notion.site/System-Overview-Core-Concepts-264ddf94fd14807a9556cf9e0e178627) →