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

# Cookbooks and Tutorials

> Build practical AI applications with ready-to-use examples, workflows, and integrations powered by PiyAPI.

From first memory workflows to production-ready agent systems.

Cookbooks provide practical implementation examples for building with PiyAPI. Each guide demonstrates a reusable application pattern - from simple memory workflows to full agent systems - featuring working code, architectural constraints, and integration patterns adapted directly for production.

These examples are engineered to be copied, modified, and shipped.

* More personalized AI applications
* More reliable context retrieval
* Reduced unnecessary model interactions
* Reusable memory workflows
* Practical implementation patterns

## Pick by compatibility

Every cookbook is tagged with its target environment. Filter the library to match your deployment stack.

| COOKBOOK | CATEGORY | COMPATIBILITY |
| :- | :- | :- |
| **Personalized AI Tutor** | Companion | Platform + Self-hosted |
| **Build a Coding Companion** | Companion | Self-hosted |
| **Memory-Powered Support Agent** | Operations | Platform + Self-hosted |
| **Content Creation Workflow** | Operations | Platform |
| **Multi-Session Research Agent** | Operations | Self-hosted |
| **Visual Memory Retrieval** | Multimodal | Platform + Self-hosted |

## Start here

The fastest way to build something useful with PiyAPI.

* **Build an AI companion**: Create a memory-powered assistant with a complete end-to-end lifecycle.
* **Self-host with Ollama**: Run a local AI workflow with Ollama and persistent context.
* **Partition memory by entity**: Scope memories strictly by user, agent, session, or application.

## Essentials

<CardGroup cols={2}>
  <Card title="Build a Companion with PiyAPI" icon="robot">
    Learn the complete memory lifecycle for persistent AI companions. `COMPANION`
  </Card>

  <Card title="Partition Memory by Entity" icon="layers">
    Balance personalization with strict scoped context across boundaries. `ARCHITECTURE`
  </Card>

  <Card title="Tag and Organize Memories" icon="tags">
    Utilize categories and tags to accelerate retrieval, filtering, and auditing. `ORGANIZATION`
  </Card>

  <Card title="Export Memories" icon="download">
    Back up, migrate, and audit stored memory datasets. `DATA OPS`
  </Card>
</CardGroup>

## Companion Playbooks

Build assistants that retain user preferences, goals, and historic interaction states.

<CardGroup cols={2}>
  <Card title="Interactive Memory Demo" icon="play">
    Execute persistent memory workflows inside a real application. `DEMO`
  </Card>

  <Card title="Research Assistant for YouTube" icon="search">
    Inject personalized context directly into video research and retrieval. `RESEARCH`
  </Card>

  <Card title="Voice-First AI Companion" icon="mic">
    Build a voice-first experience leveraging persistent context. `VOICE`
  </Card>

  <Card title="Personalized AI Tutor" icon="graduation-cap">
    Maintain deterministic student progress state across sessions. `EDUCATION`
  </Card>

  <Card title="Smart Travel Assistant" icon="plane">
    Persist traveler preferences and planning context. `TRAVEL`
  </Card>

  <Card title="Build a Node.js Companion" icon="brackets-curly">
    Create a strictly-typed JavaScript companion that retains user context. `SDK`
  </Card>

  <Card title="Self-Hosted AI Companion" icon="server">
    Deploy a complete memory-powered assistant onto local infrastructure. `SELF-HOSTED`
  </Card>
</CardGroup>

## Ops & Automations

Leverage persistent context to automate recurring workflows and multi-step tasks.

<CardGroup cols={2}>
  <Card title="Automated Email Intelligence" icon="mail">
    Parse and retain relevant unstructured information from email pipelines. `AUTOMATION`
  </Card>

  <Card title="Content Creation Workflow" icon="pen">
    Persist tone, style, and content guidelines as reusable context parameters. `CONTENT`
  </Card>

  <Card title="Multi-Session Research Agent" icon="search">
    Execute asynchronous research workflows across sessions without context degradation. `RESEARCH`
  </Card>

  <Card title="Memory-Powered Support Agent" icon="headset">
    Inject historic customer context directly into support agent flows. `SUPPORT`
  </Card>

  <Card title="Collaborative Task Assistant" icon="users">
    Coordinate multi-user workflows while maintaining strict context boundaries. `COLLABORATION`
  </Card>
</CardGroup>

## Integrations & Platforms

Use PiyAPI memory inside the platforms and environments that power your AI workflows.

<CardGroup cols={2}>
  <Card title="Memory-Powered Agent SDK" icon="puzzle">
    Expose callable memory tools for autonomous agent workflows. `SDK`
  </Card>

  <Card title="Memory as an API Tool" icon="webhook">
    Utilize persistent memory within strict function-calling environments. `API`
  </Card>

  <Card title="Persistent Mastra Agents" icon="agent">
    Deploy persistent agents leveraging Mastra-based architectures. `FRAMEWORK`
  </Card>

  <Card title="Healthcare Context with ADK" icon="heart-pulse">
    Maintain compliant, relevant context across isolated agent sessions. `HEALTHCARE`
  </Card>

  <Card title="Search with Personal Context" icon="magnifying-glass">
    Execute real-time search queries against persistent user profiles. `SEARCH`
  </Card>

  <Card title="Bedrock with Persistent Memory" icon="cloud">
    Integrate cloud AI infrastructure directly with PiyAPI memory stores. `CLOUD`
  </Card>
</CardGroup>

## Frameworks & Multimodal

Explore cookbooks that combine PiyAPI with agent frameworks and multimodal applications.

<CardGroup cols={2}>
  <Card title="AI Agents with Memory" icon="bot">
    Build agents with native persistent memory storage. `AGENTS`
  </Card>

  <Card title="Multi-Agent Collaboration" icon="network">
    Distribute relevant context securely across collaborating agent swarms. `MULTI-AGENT`
  </Card>

  <Card title="Visual Memory Retrieval" icon="image">
    Combine parsed visual context alongside text-based conversation history. `MULTIMODAL`
  </Card>

  <Card title="Persistent Persona" icon="user">
    Maintain deterministic persona parameters and preferences for autonomous agents. `PERSONA`
  </Card>
</CardGroup>

## Contribute a Cookbook

Have a useful PiyAPI workflow or integration? Share it with the community.

* **Cookbook Template**: Follow the standard structure for narrative, implementation steps, and end-to-end workflows.
* **Contribution Guide**: Learn how to submit your cookbook and prepare it for the repository.

<Info>
  **Build something useful with PiyAPI?**

  Share your cookbook and help other developers discover practical memory workflows.
</Info>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.