EP 079 · 111 MIN BRANDS PAY FIVE FIGURES FOR ONE WORKFLOW
AI Workflows Brands Actually Pay Five Figures For
The short version
Drew Brucker and Rory Flynn spend a full episode building node-based AI workflows in Figma Weave live, arguing brands pay five-figure fees for reusable creative systems rather than one-off images. Rory demos production tools built for SharkNinja and BarkBox — image reversioning, a batch-processing engine, and a listicle-video pipeline — using Nano Banana, Gemini, Flux, Grok and Claude via MCP. Their argument: a system encodes taste and process, so it survives model changes and can be handed to a team.
Key takeaways
- 01
Sell workflows the way Rory did at SharkNinja: solve one small, recurring problem first, then let the results sell the bigger build.
- 02
Split one LLM output into many with an asterisk-delimited array: write 'separate each prompt with an asterisk,' then feed it into an array or text-iterator node.
- 03
Lock every node before publishing a workflow as a tool, so teammates only see the fields they need to fill in, not the engineering underneath.
- 04
Keep workflows narrow: a single-use-case tool with one photo and one system prompt beats one giant do-everything graph that breaks when one node fails.
- 05
Paste a whole Weave workflow into Claude or ChatGPT — it reads as code — to debug broken connections or ask for a tighter system prompt.
- 06
BarkBox turns 2D vector toy art into realistic 3D renders with one system prompt: same seams, same materials, nine angles, every time.
- 07
A workflow is equity: because it survives model swaps, like nano banana to nano banana pro, the build work compounds instead of resetting.
Receipts · said on the record
“Consistency, baby. Consistency at scale is the product. That's the thing that brands pay for.”
“Each one of these workflows that you may build is equity — you're building things that you can leverage for the future.”
“What I see a lot of most of the time, people try to like way over engineer, like get these giant node strings, and it's like, well, one thing breaks in the middle and then the rest of it doesn't work.”
“They have hundreds of products. They have 25 categories, 35 global markets … They also launch 25 new products a year and they do 300 unique videos every year of pieces of content. That's insane.”
“Brands pay some money for this too, guys. This is five figure stuff. This matters.”
Chapters
- 0:00:00 Cold opener
- 0:07:22 Why a full Figma Weave episode
- 0:09:53 How node workflows went mainstream in a year
- 0:14:10 Why use a node-based AI tool at all?
- 0:19:02 Inside the Figma Weave AI toolbox
- 0:20:53 How to sell AI workflows to businesses
- 0:23:24 Why brands pay five figures for AI workflows
- 0:24:33 How SharkNinja localizes across 35 markets
- 0:30:07 Why recurring tasks belong in workflows
- 0:31:34 Live build: your first AI workflow
- 0:39:09 What the Gen Effect node can do
- 0:44:00 How to turn an image into a 3D asset
- 0:48:46 Scale one fashion image into reusable assets
- 0:55:18 How to build a batch-processing engine
- 1:03:24 Prompt vs system prompt: what's the difference?
- 1:05:38 Turn a node graph into a reusable tool
- 1:12:17 Engineering an automated listicle video
- 1:14:51 One-shot video vs modular AI production
- 1:24:29 How BarkBox fixed one production bottleneck
- 1:35:20 Where AI workflows create business ROI
- 1:38:20 Why workflows survive every new AI model
- 1:39:29 Paste your workflow into Claude to debug it
- 1:42:57 Use AI workflows as callable functions
- 1:44:28 Publish your workflows into Claude MCP
- 1:46:09 Resources and wrap-up
Questions this episode answers
What is Figma Weave and why did Fast Hours dedicate a full episode to it?
Figma Weave is the node-based AI workflow tool Rory Flynn uses for production work. Drew Brucker and Rory Flynn built a full episode around it after listeners asked for a deeper dive following a brief demo weeks earlier.
How do the Fast Hours hosts sell AI workflows to brands?
Rory says he starts small and simple: he gave away free workflows, like a one-click asset-resizing tool, and when clients saw the results they came back asking for more, leading to five-figure engagements with brands like SharkNinja.
What is the difference between a prompt and a system prompt in Figma Weave?
Rory explains a prompt is the one-time request fed to an LLM, the same as typing into ChatGPT, while a system prompt is the standing rule set that governs how the model should respond every time, controlling format, structure and tone.
How did Fast Hours use Figma Weave for BarkBox and SharkNinja?
For SharkNinja, Rory's team used it to retouch old product photography and localize e-commerce assets across 35 global markets. For BarkBox, they built a compositor-plus-retexture system that turns flat 2D toy art into realistic studio renders for its monthly subscription box.
Can Claude use Figma Weave workflows directly?
Yes. Once a Weave workflow has output nodes and is published as a tool, it shows up inside Claude's MCP tools, so Claude can call it directly as a callable function instead of a person operating the canvas.
Show notes
Brands pay five figures for AI workflows that turn repetitive creative work into reusable systems teams can run at scale.
Drew Brucker and Rory Flynn speedrun their way into a useful business model while trying to explain node-based AI without melting anyone’s brain. The opportunity itself is practical. Find expensive, repetitive creative bottlenecks, turn them into reusable workflows, and make the complicated machinery simple enough that an entire team can actually use it.
The boys break down Figma Weave, node-based AI workflows, system prompts, batch processing, arrays, Claude, ChatGPT, Gemini, Flux, Nano Banana, Gen Effect, and MCP. SharkNinja and BarkBox examples show how AI workflows support localization, product visualization, asset re-versioning, creative production, team handoff, and scalable content generation.
The bigger idea here is that the value comes from encoding the process, creative logic, and standards into a system that can be reused across hundreds of assets without starting from zero.
