# Overview

## Use Cases

Ocean Network is a distributed supercomputer. The tutorials in this section are real, runnable workloads, not toy examples, designed to show you what the network can do and let you experience it hands-on.

Each tutorial ships as a **Docker container** you open in Ocean Orchestrator, point at a node selected from the **Dashboard**, and run. The algorithm executes on the remote hardware; only the results come back to you. No cloud accounts, no server provisioning, no per-seat licenses.

***

### What You Can Run Today

| Tutorial                    | Hardware | What it demonstrates                                           |
| --------------------------- | -------- | -------------------------------------------------------------- |
| Autoresearch: Autonomous ML | H200 GPU | A self-improving ML agent running for hours on a single GPU    |
| LLM Fine-Tuning             | H200 GPU | Fine-tuning encoder and decoder language models with LoRA/PEFT |
| Transformer Foundations     | H200 GPU | Hands-on attention visualization and embedding analysis        |
| Unsupervised Clustering     | H200 GPU | Seven clustering algorithms benchmarked end-to-end             |
| Data Preprocessing & EDA    | H200 GPU | Full data cleaning and exploratory analysis pipeline           |

***

### How to Run Any Tutorial

All tutorials follow the same flow using **Ocean Orchestrator** (the VS Code extension):

1. **Install Ocean Orchestrator** from the [VS Code Marketplace](https://marketplace.visualstudio.com/items?itemName=OceanProtocol.ocean-orchestrator) (also works in Cursor, Windsurf, Antigravity).
2. **Clone the tutorials repo** — [github.com/oceanprotocol/oncompute-tutorials](https://github.com/oceanprotocol/oncompute-tutorials).
3. **Open the tutorial folder** in Ocean Orchestrator using the folder selector.
4. **Browse the node catalog** at [dashboard.oncompute.ai](https://dashboard.oncompute.ai/) and select a node that matches the hardware requirements listed in each tutorial.
5. **Start the job** — click **Run Job** using a configured a paid environment for full runs.
6. **Collect results** — outputs and logs download automatically to your `results/` folder when the job completes.

See the Ocean Orchestrator guide for a full walkthrough.

***

All tutorial source code lives at [github.com/oceanprotocol/oncompute-tutorials](https://github.com/oceanprotocol/oncompute-tutorials).


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.oncompute.ai/use-cases/quickstart.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.