The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Claude Fable 5.1 is moments away from release.
You need to start preparing immediately
I HIGHLY recommend everyone builds their own testing harness based on the tasks they do most often
Trust NO benchmarks Anthropic or random people on X use
I built my own. Here's how:
Go to Claude Code (or your favorite coding agent)
Ask it to categorize the work you do into 5 high level categories
Say you want to build a model benchmarking harness. Ask it to come up with a benchmark test for each of your categories. These should be complex, multistep tests
Ask it to build a web app to host these tests. Ask it to plug into Vercel's AI router so you can get a full list of models available and test any of them head to head
The moment the model drops, run the battery of tests head to head with the new model and also compared to the model you've been using lately.
You'll be in a good spot to make a true judgement on just how good these new models are
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Challenges
View all
Related posts
I designed this mobile app in Figma, then took it a step further by building it with Claude Code.
These are some of the interactions I’ve built so far 👀
I’ve been getting much more comfortable with turning my designs into working prototypes.
Looks cool!
One of our clients needed 70 ad video variants in a week, so we built an AI agent to do it in seven hours.
The build itself was a project. It required:
- Assembling a Claude agent connected to Remotion
- Building a knowledge base of the client's brand: brand guide, movement language, hook creation guidelines, past deliverables, briefing structure
- Teaching the agent to read the brief the way our team would
- Composing hook variants adapted for each aspect ratio (1x1, 9x16, 16x9)
- Assembling endcards and outputting files named and organized for QA
The first batch still needed heavy manual editing. The second, the agent assembled alone.
The third came out fully stitched (hook, body, CTA, disclaimer) needing only a review pass before shipping.
Each iteration meant testing what the agent got wrong, adjusting the skills, refining the knowledge base, and rerunning until the output matched the standard our team would ship.
When we sent the batch, the client feedback was: "The revised videos look great. We'll start using them immediately."
To show what this workflow can do, the video in this post was produced by this exact system.
If your business needs to ship ads at this volume, let's talk 👉 apta.agency/contact
70 ad variants in seven hours is wild. remotion?
Practical diagnostic
Before buying another tool, ask one question:
Where does the work actually stall?
Not where the dashboard says it stalls. Where people wait, re-enter information, hunt for context, or chase an approval.
That answer usually matters more than a long software wish list.
My first step on an automation project is to map the real handoffs. The goal is not to automate everything. It is to remove avoidable work without creating a new mess.
If you can describe the process in five sentences, I can usually spot the first useful improvement.
#BusinessSystems #ProcessImprovement