Fast Hours

Creative teams get to one click by building and testing Figma Weave workflows, then packaging repeat work into shared tools.
Figma Weave Creative Workflow Advocate Ron Baranov shows how the hours behind a one-click AI workflow become a tool the whole team can run. Drew Brucker and Rory Flynn follow the builds, from looping animations to locked illustration styles, and Rory discovers version history after explaining his backup routine. (Excellent timing.)
[Discussed in this episode]
Figma Weave connects large language models (LLMs), image generation, and video generation into repeatable creative workflows. Ron Baranov covers automated prompt writing, looping animations, locked brand styles, character contact sheets, and shared team tools. Figma integration, Model Context Protocol (MCP), version history, and Gen Effect controls support production workflows and creative decisions.
#FigmaWeave #AIWorkflows #CreativeAutomation #FastHours
---
⏱️ Fast Hour
00:00 Opening and Ron Baranov's Creative Background
08:07 What a Creative Workflow Advocate Does
12:39 Helping Enterprise Teams Build AI Workflows
17:11 The LLM Gap in Creative Workflows
24:35 Learn Figma Weave by Taking Workflows Apart
27:24 One Prompt, Six Image Models
32:04 LLMs Write Camera-Angle Prompts
34:44 Which AI Models Can Enterprise Teams Use?
38:30 Teaching AI Without Overwhelming a Team
43:24 Automating Loop Animations
45:35 Turning a Workflow Into a Tool the Team Can Run
47:27 A Brand Illustration Tool With One Input
49:35 Connecting Weave Assets to Figma Frames
55:17 Running Weave Tools Through Claude and MCP
57:30 What It Takes to Get to One Click
1:02:01 Using Custom Weave Tools Inside Figma
1:04:47 Building Characters From Visual Contact Sheets
1:07:28 Version History Surprises Rory
1:11:36 Keeping Creative Choices in the Workflow
1:14:18 Inside the Workflow Advocate Team
1:18:45 Creating Custom Effects With Gen Effect
1:23:17 Getting a Whole Creative Team Using AI
1:31:28 The Creative Skills Ron Would Learn Next
1:34:49 Behind the Figma Config Presentations
1:39:33 Figma Weave Office Hours and Community
1:40:13 Handlers, Fixers, and the Closing Shout-Outs
---
✱ Resources from the episode
/ Workflows & Tools /
Ron's create character workflow: https://app.weavy.ai/flow/ChkZWxAE74RbiTn20eIjbc
Intro to Figma Weave workflow: https://app.weavy.ai/flow/2ACz7n2ycDm6jxbh0D0jf1
Gen Effect workflow: https://app.weavy.ai/flow/Y5IBkzwFyc0zeYioDmGB6K
Extract a product from image tool: https://app.weavy.ai/flow/YI6F6lHjCn7K6T6HZg1bjg
/ Tutorials /
Weave first steps: https://www.youtube.com/watch?v=1pwXwbtqjBQ&t=1426s
Figma Weave youtube channel: https://www.youtube.com/channel/UCWqrnJYussMFHy5_AxwK5gg
/ Ron's Social /
Linkedin: https://www.linkedin.com/in/ron-baranov-b31740100/
Instagram: https://www.instagram.com/ronbaranov/
---
⚡️ Explore all episodes, transcripts, tools, and guests: https://fasthourspodcast.com/

Drew Brucker and Rory Flynn have officially reached the point where their AI agents are doing more work than they can keep track of.Six months ago, they were discussing who they'd need to hire to scale Fast Hours. Now they've built an interconnected system handling everything from podcast clips and thumbnails to website publishing, creative production, and client workflows.Naturally, that freed up enough time for Drew to spend an unreasonable number of hours building a squirrel video game with his dad, while Rory assembled an army of AI agents inspired by Lord of the Rings.Somewhere between the squirrels, the robots, and the existential questions about creative work, there's a pretty compelling picture of what happens when AI gives people the ability to execute their ideas.[Discussed in episode]Drew Brucker and Rory Flynn explain AI agents, Claude skills, Model Context Protocol integrations, and reusable creative workflows. Fast Hours production covers Riverside, Cloudflare, YouTube packaging, clip selection, and thumbnails. Creative experiments span Blender, Magnific, Runway, Higgsfield, and Figma. Examples show agent teams, model specialization, automated quality control, and context-driven production.#AIAgents #CreativeAI #AIWorkflows #ContentCreation #FastHours---⏱️ Fast Hour00:00 Fast Hours hits 100K subscribers04:21 Flying cars are getting ridiculous11:26 Cars are becoming software platforms14:40 Why AI is destroying traditional software moats16:09 AI can reverse-engineer entire software products20:32 The competitive advantages that still matter23:11 We built an entire creative team with AI agents25:33 The AI creative explosion30:26 Combining AI, Blender, Excel, and code32:25 The new formula for creating with AI34:37 Building a video game with Claude43:13 Turning curiosity into an AI learning program46:44 A new approach to creative ideation48:41 Building a team of specialized AI agents52:39 Tracking how AI crawlers interact with websites54:51 Using AI skeptics to improve creative work56:47 Our AI-powered podcast production system1:00:51 Cutting thumbnail production from 6 hours to 30 minutes1:02:37 Automating podcast publishing with AI agents1:04:45 Building reusable AI workflows for clients1:07:01 Running entire creative projects with AI agents1:09:49 Creating interactive websites with AI1:12:22 Automated creative testing and quality control1:17:10 Better AI video with virtual camera operators1:18:39 Recreating Fast & Furious scenes with AI and Blender1:21:51 The problem with unlimited creative possibilities1:23:36 Drew shares 12 free Claude skills1:26:18 Rory's Lord of the Rings AI agent army1:29:24 Should AI agents handle your personal life?1:36:03 Midjourney explores MCP integration1:36:55 Apple, Meta, and the future of local AI1:41:38 Remembering Napster, LimeWire, and early computing1:46:43 Final thoughts and shout-outs---🔗 ResourcesDrew's Claude Skills: https://github.com/dcbruck/claude-skillsDrew's Squirrel Game: https://squirrelscramble.com/Fast Hours Podcast Website: https://fasthourspodcast.com

Phil Rivers is using AI to scale his agency by automating approvals, task creation, onboarding, and shared workflows across his team.Drew and Rory spend an embarrassing amount of time obsessing over AI visuals. Phil Rivers shows up with 40 clients, 16 employees, and an entirely different obsession: automating all the boring stuff nobody wants to do.And it turns out the boring stuff might be where the biggest opportunities are hiding.Phil has rebuilt his agency around an AI-powered operating system that handles client approvals, project management, task creation, onboarding, and internal knowledge sharing.His goal? Double the number of clients his existing team can handle without hiring another employee.And he's already built much of the infrastructure to pursue it.Phil Rivers breaks down an AI-powered agency operating system built with Claude Code, Model Context Protocol (MCP), Supabase, GitHub, Slack, Asana, and Figma. Automated client approvals create campaign tasks and deadlines. A Slack agent named Tim drafts Asana tasks with dependencies. Klaviyo data, onboarding automation, permissions, and shared knowledge help teams handle more clients.#AIAutomation #ClaudeCode #AgencyOperations #FastHours
---
⏱️ Fast Hour
00:00 Who is Phil Rivers?
08:16 Why rebuild an agency with AI?
11:15 How an email agency first used Midjourney
14:24 How do you adopt AI without slowing down?
24:35 How to build an agency on AI from day one
27:03 How can Asana automate campaign planning?
34:36 Why service keeps clients longer than results
37:30 How do you build client context into AI?
43:20 Who decides what the AI remembers
47:38 How AI unblocks account leads to just serve
49:52 How can teams share AI skills?
54:11 How does one MCP support an entire team?
58:17 Why one-off tasks break agency teams
1:00:13 Meet Tim, the Slack agent that writes tasks
1:05:41 How many clients can AI add with zero new hires
1:09:19 Can AI shorten employee onboarding?
1:14:27 Which AI projects aren't worth building?
1:18:20 How do you keep AI workflows current?
1:19:13 How to prove AI ROI to your boss
1:24:50 How do you secure shared AI infrastructure?
1:30:11 What can Obello automate for brands?
1:34:42 How many clients can the team handle?
1:35:42 How does AI automate client onboarding?
1:38:33 What's Phil's biggest AI operations lesson?
1:41:24 Who should viewers tell about Fast Hours?

Claude Opus feels like Midjourney because it can take minimal direction and produce imaginative visual work.Drew Brucker and Rory Flynn gave Claude Opus room to create, then watched the to-do list get ambitious (their neglected websites finally got some attention). A promo’s visual style carries into animated podcast clips, while a handwriting experiment reshapes Drew’s site.Claude Opus powers motion graphics, podcast clipping, and website builds from creative references. Midjourney supplies the creative comparison and visual references for animation. Riverside Model Context Protocol connects recordings and transcripts. ChatGPT image generation builds animation assets. Magnific and ElevenLabs support sound effects and narration.#ClaudeOpus #Midjourney #MotionDesign #FastHours---⏱️ Fast Hour00:00 What's coming in Fast Hours 85?04:20 Why does context improve creative quality?05:35 Why does Claude Opus 5.5 feel like Midjourney?06:07 One design system, a creative domino effect09:07 We built a custom AI clipping workflow10:01 A blind grader judging clip hooks?13:35 Why does Riverside MCP matter for clipping?14:26 How do reference clips train an AI editor?16:37 How do you build clips around out-of-order hooks?19:32 Podcast clips get branded motion graphics23:52 Motion designers losing their sh*t over AI democratization29:58 How precise are code-based motion graphics edits?32:14 Domain expertise raises the ceiling for AI motion design39:17 Drew let his daughters' handwriting shape his brand website44:52 How do you prompt a cohesive website redesign?49:56 Fast Hours builds a website from its creative DNA52:05 Can transcripts power a searchable website?58:28 Music-reactive web design with a face1:00:20 Code-based animation gets a story to tell1:04:00 How does one image become an animation asset library?1:05:48 How do AI tools add animation and narration?1:11:24 How does ElevenLabs find a voice from a description?1:16:15 How real does a Tavus conversation feel?1:21:41 What will Phil Rivers teach about AI operations?

Opus 5.5 made editable AI video practical: build motion graphics in code, then revise shots, text, and variations without starting over.A cocktail graphic from Rory’s T-shirt became an animation system. Then came four more drinks, a Pokémon battle, podcast promos, and a client-style video. Drew and Rory sound like two people who’ve spent years hearing “one tiny change” and knowing exactly how much work that used to mean.Claude Opus 5.5 turns one cocktail image into a JavaScript or HTML animation system and produces four more recipe videos. Three.js lets Rory block a Pokémon battle and change camera angles for complex scenes. Riverside supplies context for Fast Hours promos. Drew tests a client-style video using a website and company context file.#Opus55 #AIVideo #MotionGraphics #claudeai ---⏱️ Fast Hour00:00 What are Drew and Rory making this week?07:31 Why is AI adding work for project managers?9:29 Why most companies are stuck at basic AI17:36 What is production debt, and why AI took off20:45 Is the $200 AI plan a different product?21:19 What changed with Claude Opus 5.525:04 How one T-shirt image became a motion graphic26:28 Why code-rendered video is infinitely editable28:19 How to turn one render into an animation pipeline31:25 Can Opus 5.5 render a 3D Pokémon battle?36:58 How can Three.js block a complex shot?41:36 How did AI make a Fast Hours promo?42:28 Do you need brand guidelines for this?45:41 Can AI pick a stronger clip hook?47:34 Cutting a promo with no instructions49:18 Why creatives aren't siloed anymore50:39 Watching a sponsor promo render live53:38 What one Opus 5.5 prompt gets you for a B2B SaaS video56:09 The exact prompt behind the one-shot video57:21 Claude Code vs Cowork for motion builds59:17 How do editing and motion graphics combine?1:03:05 How did Drew turn his kids' handwriting into a font?1:10:33 Stop sending AI output without the context1:14:17 How to share a skill that runs anywhere1:18:40 Why trading transcripts beats trading drafts1:20:40 Can you share a Claude project?1:24:18 How to get email discounts, then unsubscribe1:27:44 Where does the episode close?