---
schema: 1
type: video
title: 'GPU Cloud Deployment Without Leaving Your IDE — Audry Hsu, RunPod'
description: >-
  The iteration cycle before Flash: commit, push, build a Docker image, pull it
  from the registry, load it onto a server, allocate a GPU, then find out if it
  works. Audrey Hsu demos what replacing that with a single decorator looks like
  — add `@flash.endpoint` to an async Python function and it deploys to GPU
  cloud from your IDE, with hot reload so a model swap is one line of code
  rather than a container rebuild. The second demo chains three models: Qwen 3
  generates image prompts, DreamShaper renders them, Nano Banana 2 composes the
  results into a single photo. H100 pricing is $0.00116 per second, charged only
  while a worker is handling a request. RunPod's recommendation: start with pods
  while experimenting, switch to serverless when you need hundreds of workers
  autoscaling across data centers. Speaker info: -
  https://www.linkedin.com/in/audry-hsu/
canonical: 'https://liquidgold.cc/watch/zDGHt0LB-dA'
markdown: 'https://liquidgold.cc/watch/zDGHt0LB-dA.md'
llms: 'https://liquidgold.cc/watch/zDGHt0LB-dA/llms.txt'
parent: 'https://liquidgold.cc/channel/aidotengineer.md'
channel: aidotengineer
updated: '2026-10-24T09:46:31.301Z'
language: null
children: []
videoId: zDGHt0LB-dA
durationSeconds: 1219
publishedAt: '2026-06-09T18:15:10.000Z'
searchChunks:
  - ordinal: 0
    text: null
    startSeconds: 26
    endSeconds: 150
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=26'
    heldReason: rights
  - ordinal: 1
    text: null
    startSeconds: 26
    endSeconds: 150
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=26'
    heldReason: rights
  - ordinal: 2
    text: null
    startSeconds: 160
    endSeconds: 258
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=160'
    heldReason: rights
  - ordinal: 3
    text: null
    startSeconds: 258
    endSeconds: 270
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=258'
    heldReason: rights
  - ordinal: 4
    text: null
    startSeconds: 270
    endSeconds: 400
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=270'
    heldReason: rights
  - ordinal: 5
    text: null
    startSeconds: 270
    endSeconds: 400
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=270'
    heldReason: rights
  - ordinal: 6
    text: null
    startSeconds: 270
    endSeconds: 400
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=270'
    heldReason: rights
  - ordinal: 7
    text: null
    startSeconds: 410
    endSeconds: 514
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=410'
    heldReason: rights
  - ordinal: 8
    text: null
    startSeconds: 410
    endSeconds: 514
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=410'
    heldReason: rights
  - ordinal: 9
    text: null
    startSeconds: 410
    endSeconds: 514
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=410'
    heldReason: rights
  - ordinal: 10
    text: null
    startSeconds: 514
    endSeconds: 526
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=514'
    heldReason: rights
  - ordinal: 11
    text: null
    startSeconds: 658
    endSeconds: 741
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=658'
    heldReason: rights
  - ordinal: 12
    text: null
    startSeconds: 752
    endSeconds: 831
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=752'
    heldReason: rights
  - ordinal: 13
    text: null
    startSeconds: 841
    endSeconds: 954
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=841'
    heldReason: rights
  - ordinal: 14
    text: null
    startSeconds: 841
    endSeconds: 954
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=841'
    heldReason: rights
  - ordinal: 15
    text: null
    startSeconds: 841
    endSeconds: 954
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=841'
    heldReason: rights
  - ordinal: 16
    text: null
    startSeconds: 967
    endSeconds: 1077
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=967'
    heldReason: rights
  - ordinal: 17
    text: null
    startSeconds: 1077
    endSeconds: 1088
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=1077'
    heldReason: rights
  - ordinal: 18
    text: null
    startSeconds: 1088
    endSeconds: 1219
    watch: 'https://liquidgold.cc/watch/zDGHt0LB-dA?t=1088'
    heldReason: rights
---


# GPU Cloud Deployment Without Leaving Your IDE — Audry Hsu, RunPod

The iteration cycle before Flash: commit, push, build a Docker image, pull it from the registry, load it onto a server, allocate a GPU, then find out if it works. Audrey Hsu demos what replacing that with a single decorator looks like — add \`@flash.endpoint\` to an async Python function and it deploys to GPU cloud from your IDE, with hot reload so a model swap is one line of code rather than a container rebuild. The second demo chains three models: Qwen 3 generates image prompts, DreamShaper renders them, Nano Banana 2 composes the results into a single photo. H100 pricing is $0.00116 per second, charged only while a worker is handling a request. RunPod's recommendation: start with pods while experimenting, switch to serverless when you need hundreds of workers autoscaling across data centers. Speaker info: - https://www.linkedin.com/in/audry-hsu/

Canonical: https://liquidgold.cc/watch/zDGHt0LB-dA · Channel: [AI Engineer](https://liquidgold.cc/channel/aidotengineer.md)

> Search chunk text is held until public-source provenance is verified.

## Document chunks

> Missing: no stored document chunks are available.