📚 Learning Flow

AI Agent Tutorial

Quick Start (3 minutes)

System Overview & Core Concepts

Run in Colab (No Code)

Build Your Own AI Agent

FAQ / Troubleshooting

Purpose of This Tutorial

Welcome! This guide is for professors and researchers from any discipline—with or without an engineering background—who want to explore the Aralia ecosystem. With the Aralia AI Agent, you will:

1. Search suitable datasets across multiple Data Planets (e.g., SDG, UN).
2. Align time ranges and geographies to produce a comparable table.
3. Summarize findings with an LLM into a concise, readable insight.

By the end, you can run a complete workflow in Colab without writing code. If you’d like to go further, you’ll also understand the five Agent stages and the LangGraph flow so you can integrate the process into teaching or research.

Friendly reminder: the clearer you state your

research/decision question

Aralia AI Agent

What You’ll Be Able to Do

• Converse to insights

  Ask a well-formed question and let the AI Agent retrieve and align data, then return an answer with sources and limitations.

• No-Code / Low-Code

  Use our prepared Colab environment to run the full workflow with no code, or use low code to adjust a few parameters or customize the prompt.

• Course-ready

  Add your question to a sample question bank and re-run with one click for classroom use and reproducibility.

• Start your Agent journey

  Take your first steps toward building your own AI Agent–powered service.

What You’ll Build

• Minimum Viable Example (Malaysia: GDP × Gini)

  Question:

  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?

  What the Agent does:

  Finds GDP and Gini datasets across SDG/UN Planets → pulls 2021–2024 and 2024 data → aligns by state → returns a table plus a short insight.

• Your own case (pick from the Sample Question Bank)

  Copy a complete question into Colab and run it as-is, or swap in your country/year to match your topic.

Tools & Resources

• Colab Notebook

  Includes environment setup, Secrets (LLM API Key, OAuth), one-click run, and result viewing.

• Sample Question Bank

  A curated set of copy-ready questions to explore real scenarios.

• Aralia OpenRAG Library (optional, advanced)

  Learn the 5 Agent stages and the LangGraph flow. Advanced users can tweak interpretation_prompt or add custom nodes in the SDK.

How It Works (High-Level Overview)

• Five Agent Stages (orchestrated with LangGraph):

  1. Dataset Search (semantic retrieval)
  2. Analytics Planning (decide x / y / filter and field formats)
  3. Filter Decision (finalize years/geo into executable parameters)
  4. Analytics Execution (query Planet APIs; align time & geography)
  5. Interpretation (LLM writes a ≤300-word insight with constraints)

• LLM selection

  The notebook picks a provider based on your API key (Gemini / OpenAI / Anthropic).

• RAG & vectorization

  Retrieval and embeddings are handled by the system. Your job is to ask a clear, fully specified question (metric, time, geography, relationship).

Ready to go? Let’s begin with the quick demo and build momentum!

                                                                                           Next： [

Quick Start (3 minutes)](https://deciduous-centipede-9d7.notion.site/Quick-Start-3-minutes-269ddf94fd148196aba0ecafdb719b3c) →