
Product Engineer Building @Studentcornr Member @SuperteamNG Javascript Buff Everything code and software 😃
The web
Joined March 2022
- Tweets1.4K
- Following407
- Followers118
- Likes4.1K
>be me
>used to think @poteto's skills were a meme
>decided to try them for fun
>set them up across all my repos
>fed matt x poteto podcast transcript and asked codex to make my workflows better
>200+ PRs merged to main in the last 7 days 😭😭😭
>credit goes to tibo's usage resets too lol
The update came very early than expected.
I just checked through my mail now quite long now 😂
Unto the next wish me luck 🤝
I had a conversation with an HR from Poland recently, and the call went really well.
Towards the end, she casually asked, “You’re from what part of Africa again?”
I said Nigeria, and she went, “Ooooh, brilliant!”
Now I’m wondering if I’ve got a Nigerian predecessor who left a good impression or if I’m just overthinking things. 😂
I’ll share an update in the coming days or weeks.
Introducing Jobmate (@usejobmate): the AI copilot that job hunts with you, and for you, across your entire job search.
It’s a brand new way to find jobs, auto-apply (with a humanized agent that runs on the web), autofill forms, find leads or reach hiring managers, build resumes from successful people’s profiles, and track your entire job search in one place.
If you are currently job hunting, this would be the best 2 mins of your day. If you aren’t job hunting right now, share this with a friend or your network.
Try it at usejobmate.com
new skill: create-canvas.
last week I made a skill to put every component from a repo on one canvas. I started using it with my agent to create variations, compare ideas, and iterate visually without leaving the codebase.
so I made a new skill around that workflow.
instead of moving your work into another tool, bring the canvas into your codebase and use it to explore UI with your agent, using the real components, styles, and project context already there.
When preparing a meal, it is important to know what to add BUT what is just as important is WHAT NOT TO ADD because it makes a huge difference.
Stay cooking! 🧑🍳
Hi @shadcn the shadcn linter has gap, it's not able to recognize custom font sizes declared in css theme, it assumes the `text-` token is only for colors, so when you declare a custom text size in the theme and use it in your app, the `no-raw-colors` flags it.
⚠️ Frontier AI models still fall significantly short of human designers on design creativity. ⚠️
Introducing Design Creativity Bench, a new benchmark evaluating AI models’ creativity on UI design tasks.
Despite rapid advances in their ability to generate functional websites and product UIs, leading AI models still struggle to make creative design decisions.
We evaluated 16 frontier models across 80 open-ended UI tasks, each adapted to two distinct product domains alongside human designer references.
Existing UI benchmarks often measure a single output quality metric. Instead, we measure three dimensions of design creativity:
- Originality: How distinct a model's designs are from other models' outputs for the same brief.
- Creative Range: How much a model adapts its design decisions when the product domain changes.
- Appropriateness: How well the generated UI satisfies the functional requirements of the brief.
The results reveal a substantial gap in overall creativity: The best model, @OpenAI GPT-6 Astra, scores 65.5 against a human reference of 92.3.
We're publishing the full leaderboard, methodology, and reproducible evaluation framework to make design creativity a measurable dimension of frontier model capability.
📊 Benchmark Leaderboard: kombai.com/benchmark/
📄 Paper: arxiv.org/abs/2610.11539
🤗 HuggingFace Datasets: huggingface.co/datasets/Komb…
🔗 Github: github.com/kombai-io/design-…
"How do u even know what good looks like?"
This has been my qualm with Agentic coding as a junior, AI spits out something & I just review it with other 3 agents & merge it.
Till last year it was thorough discussions within team on design & I used to learn a lot about stuff by just being there; now it's, "Just ask claude & see what it says."
Those team discussions, design stuff & nerding out on topics has been lost, idk about others but I've experienced it.
So, I'm now learning through building by hand & AI on the side to discuss, but seeing everyone building autonomously with agents also makes me feel left behind alot.
The thing is idk enough about things, so I can't rely on agents to build for me & expect them to be good at what they are outputting, all I can do is review with agents & test the output.
Even if i ask AI what it did, reading through it feels cheap & I forget it anyways, coz I didn't write it.
Nobody wants me to learn, they want me to deliver & ship like a worker on an assembly line.
@poteto Could you chime in on this as someone who transitioned to a senior before AI. What do you recommend to juniors in the AI era, how do they know what "good" is in the AI era.
True, but there are certain fields that are more AI-proof than others.
I've authorized Kombai for a project and we didn't even need a UI designer or frontend developer.
Now on the other hand, this cannot be done on something like cloud and infrastructure operations for instance.
There's something ridiculous about a developer in Lagos paying Silicon Valley subscription prices just to experiment with coding agents.
We should be building alternatives that account for local purchasing power.
Asking Claude to review your code is a slot machine.
Every review finds brand new bugs. Review again: more. Review a tenth time: it still finds something.
It never runs out of bugs to find... so how do you decide your code is actually done?
I added a new /state-machine skill to the collection.
It renders any component you want in all of its states on a throwaway page with mock data.
I enjoy working like this a lot. It makes it very easy to work on more complex components.
github.com/jakubkrehel/skill…
I spent the majority of my career building observability tooling
this tooling should be used to close the loop between production and coding agents, in service of self-improving software
I put my thoughts down
boristane.com/blog/self-impr…






















