# Temporal

> Explore Melanie Warrick’s Temporal demos, Vibe Check episodes, and resources for building AI agents with durable execution.

- Author: Nyghtowl (Melanie Warrick)
- URL: https://nyghtowl.com/temporal/
- Published: 2026-05-03
- Site index for agents: https://nyghtowl.com/llms.txt


[Explore the demos](https://nyghtowl.com/temporal/#demos)



**Melanie** <img src="https://nyghtowl.com/images/nyghtowl100.png" alt="nyghtowl owl logo" style="display: inline; height: 1.4em; vertical-align: -0.35em; margin: 0 0.1em;"> I build and demo AI agents with durable execution at [Temporal](https://temporal.io?utm_source=nyghtowl.com&utm_medium=referral&utm_campaign=temporal-page&utm_content=intro). Find me on [LinkedIn](https://www.linkedin.com/in/melaniewarrick) and [GitHub](https://github.com/nyghtowl).


[Start here](#start-here) · [Demos](#demos) · [Vibe Check](#vibe-check-series) · [Resources](#resources)

**Latest Vibe Check:** [What survives when the agent session ends?](https://youtu.be/FPhLBODA2LY)

## Start here

Temporal is an open-source platform for durable execution. Your code survives crashes, outages and restarts, which makes it a good fit for AI agents running long, multi-step, failure-prone work like LLM calls, tool use and human-in-the-loop approvals.

- **Install the CLI**: `brew install temporal` then `temporal server start-dev`
- **Add the agent skills** to any AI coding agent (Claude Code, Codex, Cline): `npx skills add https://github.com/temporalio/skill-temporal-developer`
- **Docs and downloads**: [temporal.io](https://temporal.io?utm_source=nyghtowl.com&utm_medium=referral&utm_campaign=temporal-page&utm_content=start-here)

## Demos

[Agent Memory & State](#agent-memory--state) · [Durable Research Fleet](#durable-research-fleet) · [Rust Confessional](#rust-confessional) · [Durable HITL Agents](#durable-human-in-the-loop-agents) · [ADK & Temporal](#adk--temporal-integration) · [Sheep Audio Dreams](#voice-demo-sheep-audio-dreams)

### Agent Memory & State

The same autonomous refund loop run twice. The naive Worker dies before calling Stripe and restarts from scratch, while the durable one reconnects to its Temporal Workflow and picks up at issue refund. No agent framework involved, which is the point.

<div style="margin: 16px 0;">
  <a href="https://github.com/temporal-community/agent-memory-and-state" target="_blank" rel="noopener">
    <img src="https://nyghtowl.com/images/agent-memory-demo-reel.gif" alt="Side-by-side: the naive agent loses its answers and restarts while the Temporal-backed agent resumes the saved loop" style="max-width: 560px; width: 100%; border-radius: 6px; display: block;">
  </a>
</div>

<div style="position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden; max-width: 560px; margin: 16px auto; border-radius: 6px;">
  <iframe src="https://www.youtube.com/embed/__zlUvJR6Hg" title="What does an AI agent need to remember? Context, memory and state in 2 minutes" style="position: absolute; top: 0; left: 0; width: 100%; height: 100%; border: 0;" allow="accelerometer; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen></iframe>
</div>

- Read the write-up: [What Does an AI Agent Need to Remember?](https://nyghtowl.com/posts/2026/09/what-does-an-agent-need-to-remember/): context, memory and state pulled apart with a refund agent and a python plushy
- Get the Code: [temporal-community/agent-memory-and-state](https://github.com/temporal-community/agent-memory-and-state)

<details class="demo">
<summary id="durable-research-fleet">Durable Research Fleet</summary>
A deep-research agent on Temporal Serverless Workers that fans a question out to three to six parallel researchers on Cloud Run and merges one cited report. The worker pool scales to zero while the Workflow stays parked waiting on your approval, then wakes when you give it.

Originally [Shubham Londhe](https://github.com/LondheShubham153)'s project. I moved it to Gemini and we demoed it together at a GDE talk.

<div style="margin: 16px 0;">
  <a href="https://github.com/temporal-community/durable-research-fleet" target="_blank" rel="noopener">
    <img src="https://nyghtowl.com/images/research-fleet-console.png" alt="Durable Research Fleet console: outline, cited report, sources, and live fleet counters" style="max-width: 560px; width: 100%; border-radius: 6px; display: block;">
  </a>
</div>

- Get the Code: [temporal-community/durable-research-fleet](https://github.com/temporal-community/durable-research-fleet)

</details>

<details class="demo">
<summary id="rust-confessional">Rust Confessional</summary>

A durable AI agent in Rust, live on stage: the audience confesses programming sins and Ferris the agent delivers judgment, keeping its place through rate limits, network partitions and a Worker crashed mid-absolution. **Wall of Regrets** is the same demo running as a passive booth display.

Co-presented with [Melissa Herrera](https://github.com/melienherrera).

<div style="margin: 16px 0;">
  <a href="https://youtu.be/_t_Rxf8Z4mU" target="_blank" rel="noopener">
    <img src="https://img.youtube.com/vi/_t_Rxf8Z4mU/maxresdefault.jpg" alt="Rust Confessional talk recording" style="max-width: 560px; width: 100%; border-radius: 6px; display: block;">
  </a>
</div>

A 10-second walkthrough: the judgment parks at its human-in-the-loop checkpoint, the Worker is replaced mid-deployment, the reply arrives during the gap, and a fresh Worker finishes the same agent.

<div style="margin: 16px 0;">
  <video controls preload="metadata" poster="/videos/wall-of-regrets-poster.png" style="max-width: 560px; width: 100%; border-radius: 6px; display: block;">
    <source src="https://nyghtowl.com/videos/wall-of-regrets-walkthrough.mp4" type="video/mp4">
  </video>
</div>

- Watch the Talk: [Rust Confessional](https://youtu.be/_t_Rxf8Z4mU)
- Slides: [Rust Can Fix That (PDF)](https://github.com/temporal-community/rust-confessional/blob/main/docs/media/rust-can-fix-that-slides.pdf)
- Get the Code: [temporal-community/rust-confessional](https://github.com/temporal-community/rust-confessional)

</details>

<details class="demo">
<summary id="durable-human-in-the-loop-agents">Durable Human-in-the-Loop Agents</summary>

Framework-agnostic human-in-the-loop patterns: agents that pause for human input and reliably pick back up, no matter how long the wait or which framework you use.

<div style="margin: 16px 0;">
  <a href="https://youtu.be/kTPDzsXxKFg" target="_blank" rel="noopener">
    <img src="https://img.youtube.com/vi/kTPDzsXxKFg/hqdefault.jpg" alt="Durable Human-in-the-Loop Agents demo video" style="max-width: 560px; width: 100%; border-radius: 6px; display: block;">
  </a>
</div>

- Follow along: [temporal-community/durable-hitl-agents](https://github.com/temporal-community/durable-hitl-agents)

</details>

<details class="demo">
<summary id="adk--temporal-integration">ADK & Temporal Integration</summary>

Google Agent Development Kit integrated with Temporal for durable, production-ready AI agents.

<div style="margin: 16px 0;">
  <a href="https://www.youtube.com/watch?v=J7WGf7YeUR4" target="_blank" rel="noopener">
    <img src="https://img.youtube.com/vi/J7WGf7YeUR4/maxresdefault.jpg" alt="ADK and Temporal integration demo video" style="max-width: 560px; width: 100%; border-radius: 6px; display: block;">
  </a>
</div>

- Get the Code: [temporal-community/ice-cream-fleet-demo](https://github.com/temporal-community/ice-cream-fleet-demo)
- Read the write-up: [Durable, flexible multi-agent systems](https://nyghtowl.com/posts/2026/08/durable-flexible-multi-agent-systems/): the same fleet on ADK, LangGraph, and both at once, with durable human-in-the-loop patterns

</details>

<details class="demo">
<summary id="voice-demo-sheep-audio-dreams">Voice Demo: Sheep Audio Dreams</summary>

A voice-driven AI agent demo built with Temporal.

<div style="margin: 16px 0;">
  <a href="https://www.youtube.com/watch?v=-mp9beFRJ1Y" target="_blank" rel="noopener">
    <img src="https://img.youtube.com/vi/-mp9beFRJ1Y/maxresdefault.jpg" alt="Sheep Audio Dreams voice demo video" style="max-width: 560px; width: 100%; border-radius: 6px; display: block;">
  </a>
</div>

- Get the Code: [temporal-community/sheep-audio-dreams](https://github.com/temporal-community/sheep-audio-dreams)
- Get the Code: [nyghtowl/sheep-audio-dreams](https://github.com/nyghtowl/sheep-audio-dreams)

</details>

## Vibe Check Series

<!-- The embed below is the latest episode. When it changes, update the Latest Vibe Check link at the top too. -->

My ongoing YouTube series exploring AI tools, agentic coding, and how to build AI in production. Watch the full **[Vibe Check playlist](https://www.youtube.com/playlist?list=PLl9kRkvFJrlREy2qVAyrh9cPe8MzhVdfX)**.

<div style="position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden; max-width: 560px; margin: 16px auto; border-radius: 6px;">
  <iframe src="https://www.youtube.com/embed/FPhLBODA2LY" title="Vibe Check: What survives when the agent session ends?" style="position: absolute; top: 0; left: 0; width: 100%; height: 100%; border: 0;" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen></iframe>
</div>

## Resources

- **[temporal.io](https://temporal.io?utm_source=nyghtowl.com&utm_medium=referral&utm_campaign=temporal-page&utm_content=resources-site)**: Official website and docs
- **[temporalio/temporal](https://github.com/temporalio/temporal)**: Main company GitHub repo
- **[temporal-community](https://github.com/temporal-community)**: Community GitHub org
- **[AI Cookbook](https://github.com/temporalio/ai-cookbook)**: Recipes for building AI agents with Temporal
- **[Agent Skills](https://docs.temporal.io/with-ai?utm_source=nyghtowl.com&utm_medium=referral&utm_campaign=temporal-page&utm_content=resources-agent-skills)**: Official Temporal skills for AI coding agents ([developer](https://github.com/temporalio/skill-temporal-developer), [cloud](https://github.com/temporalio/skill-temporal-cloud), [design](https://github.com/temporalio/skill-temporal-design))
- **[Join the Community Slack](https://t.mp/slack)**: Connect with the Temporal community