> ## Documentation Index
> Fetch the complete documentation index at: https://studio-docs.prem.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Projects Overview

> Streamline your AI development workflow with end-to-end project orchestration in Prem Studio.

<Note>
  Check out the [Get Started](/projects/get-started) guide to create your first project and see the complete workflow in action.
</Note>

## What are Projects?

Projects in Prem Studio are your **orchestration hub** for end-to-end AI model development. Instead of managing datasets, fine-tuning jobs, and evaluations separately, Projects provide a **guided workflow** that connects all these components into a cohesive development pipeline.

Think of Projects as your **AI development roadmap** — they guide you from raw data to a production-ready, evaluated model through a structured, step-by-step process.

## The Complete Project Workflow

Project Workflow: Project Setup → Dataset Creation → Fine-tuning → Metrics -> Evaluation

Every project follows this proven workflow:

<Steps>
  <Step title="Project Setup">
    Define your project name and objective.
  </Step>

  <Step title="Dataset Creation">
    Choose your path:

    * **Upload existing JSONL** datasets if you already have training data
    * **Generate synthetic datasets** from PDFs, websites, videos, or other sources

    Learn more: [Datasets Overview](/datasets/overview) | [Synthetic Data Guide](/guides/datasets/synthetic-data)
  </Step>

  <Step title="Fine-tuning">
    Train your selected model on your dataset with automated parameter optimization.

    Learn more: [Fine-tuning Overview](/finetuning/overview) | [Get Started with Fine-tuning](/finetuning/get-started)
  </Step>

  <Step title="Metrics">
    Define custom evaluation rules and metrics that align with your specific use case and quality standards.

    Learn more: [Evaluation Metrics](/evaluations/metrics) | [Writing Good Metrics](/guides/evaluation/writing-good-metrics)
  </Step>

  <Step title="Evaluation">
    Test your fine-tuned model against your defined metrics to validate performance and reliability.

    Learn more: [Evaluations Overview](/evaluations/overview) | [Get Started with Evaluations](/evaluations/get-started)
  </Step>

  <Step title="Project Complete">
    Deploy your validated model with confidence, knowing it meets your quality standards.
  </Step>

  <Step title="Iteration Loops">
    Projects are **not one-and-done** — they’re designed for iteration:

    * **Dataset Expansion** → Refine or add new samples when evaluation highlights weaknesses.
    * **More Fine-tuning Jobs** → Train new models with different parameters or approaches.
    * **New Metrics** → Add metrics to better evaluate new behaviors.
    * **Re-evaluation** → Apply the same or updated metrics to compare across models.
  </Step>
</Steps>

## Why Use Projects?

### Guided Experience

Projects eliminate guesswork by providing a clear, sequential workflow. Perfect for users new to AI development or teams wanting standardized processes.

### Seamless Integration

All components (datasets, fine-tuning, evaluation) work together seamlessly. No need to manually connect outputs from one step to inputs of the next.

# Next Step: Create a Project

<Card title="Get Started with Projects" icon="rocket" href="/projects/get-started">
  Create your first AI project with our step-by-step creation guide.
</Card>

# Deep Dive into Components

While Projects orchestrate the overall workflow, you can learn more about each component:

<CardGroup cols={3}>
  <Card title="Datasets" icon="database" href="/datasets/overview">
    Learn about data preparation and synthetic generation.
  </Card>

  <Card title="Fine-tuning" icon="robot" href="/finetuning/overview">
    Understand model training and customization.
  </Card>

  <Card title="Evaluations" icon="chart-bar" href="/evaluations/overview">
    Explore model testing and validation methods.
  </Card>
</CardGroup>