You buildAn agent workflow on your live pipeline
Applied AI for video and media
Build an ingest, logging and rough-cut workflow on your own footage, run it inside your studio pipeline, and measure the hours it gives back.
Free
8 modules
6 weeks
Intermediate
What you will learn
- Write a creative brief that a generative pipeline can actually be held to.
- Hold character, style and continuity across shots that models generate separately.
- Direct voice, music and sound, and keep a licence and consent record for each.
- Assemble, caption, grade and export to a delivery specification.
- Keep a prompt and iteration log that stands up as provenance.
- Finish with a master, two channel adaptations and a measured before and after.
Course content
Method step 1, Profile
Audience, message, format and constraints, plus what this delivery costs today.
- Audience, platform, message, format
- Constraints, references and budget
- Record the baseline on your current delivery
- Define the success measure
Method step 2, Build
Ideation and structure, with factuality and brand voice held as constraints.
- Ideation and the logline
- Narrative structure and shot intent
- Factuality and brand voice
- The script pass
Method step 2, Build
The continuity problem, solved before generation rather than during it.
- Style frames and the mood board
- Character and style consistency
- Storyboard and shot list
- The continuity bible
Method step 2, Build
Text to video, image to video, camera language, and diagnosing a bad take.
- Text to video and image to video
- Camera language and motion control
- Iteration, and reading a failure
- Asset management and naming
- The iteration log
Method step 2, Build
Direction and timing, with consent and licensing recorded as you go.
- Voice direction
- Consent and cloning rules
- Music and effects generation
- Timing and licensing
Method step 3, Deploy
Assembly through export, to a specification somebody else could check.
- Assembly and pacing
- Transitions and lip sync
- Captions and colour
- Upscaling and quality control
- Export specifications
Method step 4, Measure
Ship the versions, then read what the platform tells you about them.
- Platform versions and thumbnails
- Accessibility as a delivery requirement
- Retention signals and the performance review
- Reuse and the outcome sheet
Method step 5, Document
The paperwork that makes the work usable, and the record that closes the course.
- Releases and copyright
- Disclosure and the source ledger
- The model and tool record
- Assemble the completion record
- Publish it and book the review
Requirements
- One production workflow you own, and permission to change it.
- Your own footage, or a project you have the rights to work with.
- An edit application you already know.
- About four hours a week for six weeks.
About this course
This course is for the people who own post: editors, producers and post supervisors. It starts from the professional process rather than from a tool list, so the generative work sits inside a brief, a storyboard and a delivery specification the way it does on a real job.
Roan's published book on applied AI in media production is the spine of the subject matter, and the studio work behind it is where these workflows were first run.
The method is the one every course here runs on. You record what one part of your pipeline costs today, build the workflow on your own footage across eight guided labs, run it on a real delivery, then measure the same thing again.
The capstone is a package rather than a showreel: a creative brief, a storyboard, a prompt and iteration log, the final master, two channel adaptations, a rights and provenance sheet, and a retrospective. That set is considerably harder to imitate than a demonstration of five tools.
Common questions
It is free. The first module of every course opens instantly, and one free account unlocks the rest of that course. Everything on this site is free to use.
A course completes when your workflow runs live and you have measured it. You record a baseline in module 1, launch the workflow into a working environment, then measure the same thing again. That pairing is what makes the completion record worth sharing.
Yes. The first module of every course plays for everyone, with a free account needed only from module 2.
Name, role, organization, and a few questions about the workflow you want to improve. It takes about a minute, and it is what lets the guided labs use your own workflow as the project.
One workflow you own, permission to change it, and the tools your team already uses. Applied AI infrastructure also assumes access to a GPU cloud account, and AI starter for small business runs on everyday business tools.
A one-page record inside the course. Before you launch, you note a baseline on time, cost, or quality. After you launch, you measure again. The sheet holds both numbers and the difference between them.
Instructor
Roan Weigert
Records all five courses
Hackathon judge and AI content creator, and host of the AI Insights San Francisco podcast. Writes this curriculum and records every core lesson.
Other courses
The other four courses
Each one takes a different kind of work through the same five steps. Pick the one closest to your job.
You buildAn AI use policy your team runs onAI literacy, ethics and data compliance
A working AI policy for your team, covering what the tools may touch and who checks it.
You buildModel serving on GPU cloud
You buildOne assistant for your busiest taskAI starter for small business
A short course to one measurable first win, built for owners and operators.
Start with module 1
Creative brief and production strategy
Open to everyone, with no account. You finish it holding a creative brief and baseline.