---
title: Overview | Adaption
description: Understand what AutoScientist builds and how its iterative training loop works.
---

AutoScientist automates the research loop behind model training and alignment. You choose the outcome; AutoScientist optimizes the data and [training recipe](/autoscientist/recommended-hyperparameters/index.md), trains a model, evaluates it, and repeats until the run reaches its goal or exhausts its iterations.

Start with a dataset that has completed an [Adaptive Data](/adaptive-data/overview/index.md) run. If your source is already a training-ready prompt/completion dataset, you can instead [run on non-adapted data](/autoscientist/run-on-non-adapted-data/index.md).

![Select a dataset for AutoScientist](/auto-scientist/ASEX1.png)

## What you can build

- **Instruction datasets** pair prompts with expected completions for supervised fine-tuning.
- **Preference pairs** compare two completions to indicate which is better for preference-based training.

AutoScientist API training is supervised fine-tuning (`sft`). There is no `method` parameter on `client.autoscientist.create`.

## How the loop works

Each run cycles through four stages:

1. **[Data optimization](/autoscientist/data-augmentation/index.md)** adds optional domain-targeted and general-purpose rows to improve diversity and reduce overfitting.
2. **[Training](/autoscientist/running-autoscientist/index.md)** runs with the selected model, LoRA or full fine-tuning, and the current hyperparameter recipe.
3. **[Evaluation](/autoscientist/interpreting-results/index.md)** measures the iteration against `target_win_rate` and captures training diagnostics such as loss, learning rate, and gradient norm.
4. **Hyperparameter adjustment** analyzes the evaluation and revises the recipe for the next iteration.

The loop stops when it reaches `target_win_rate` or completes `max_iterations`. A `succeeded` status can mean either condition, so inspect `best_win_rate` after completion.

## Next steps

- [Configure data augmentation](/autoscientist/data-augmentation/index.md)
- [Create and monitor a run](/autoscientist/running-autoscientist/index.md)
- [Interpret the results](/autoscientist/interpreting-results/index.md)
- [Download the best checkpoint](/autoscientist/download-the-model/index.md)
- [Review supported models and training limits](/autoscientist/supported-models/index.md)
- Browse every request and response field in the [`create` API reference](/api/python/resources/autoscientist/methods/create/index.md)