Ali

Developer

[Ali @amohammadpoor]

@mona_biasia @figma @claudeai Love this! Clean animation and great hero section.

[Ali @amohammadpoor]

@nishimiya @vercel Appreciate the post. Feels good to know that other developers have gone through the same phase. Sent you a dm!

[rauno @raunofreiberg]

Make better things, not more things.

[Ali @amohammadpoor]

@SebCornelius This is CLEAN

[Ali @amohammadpoor]

@ryancarson Fully agree! Spent all day yesterday designing in Figma with Opus 4.8 and it was producing great results. Rarely have to go back and adjust the design.

[Ali @amohammadpoor]

@kunchenguid Yeah fully agree with your takes. Would say that it’s faster than 4.7 and produces better results more consistently. It’s not day and night but it feels better for sure.

Also, fast mode needs to be implemented for normal subscriptions as well. That would change everything.

[Ali @amohammadpoor]

Loving Opus 4.8 so far. Feels much snappier, better context and for sure better at working at longer tasks. Great at handling to-dos and reminding me of things I forgot or pushed forward.

[Ali @amohammadpoor]

@robinebers @vercel_dev Thank you!! Would love if you gave it a test. Curious to know how it works for others and what kind of improvements I can do.

Yes, native would be amazing. They already have a great Mac app so it shouldn’t be a big step.

[Ali @amohammadpoor]

@rauchg @williambout @merycodes Looks much cleaner and it's easier to scan. Props to the team.

Used my own app in the menubar for a few months and it's been great for minimizing context switching, fixing failed deploys and sending links. Notifications on every deploy and CI/CD.

https://t.co/BYlL0P9FTX

[Ali @amohammadpoor]

@williambout @merycodes Much cleaner! Easier on the eyes which makes it more scannable.

Used my own app in the menubar for a few months and it's been great for minimizing context switching, fixing failed deploys and sending links.

https://t.co/BYlL0P9FTX

[Ali @amohammadpoor]

@elvissun @vercel It's cleaner for sure but I prefer using my menubar app. Notifications on deploys, CI/CD + all links just a shortcut away.

https://t.co/BYlL0P9FTX

[Ali @amohammadpoor]

@kmelve Are you getting bad code, bad handoff, or both?
Works well when the Figma file has structure, clean naming and auto-layout (Claude can sort that out). Building component by component, not full pages, is what makes it click for me. Happy to share examples.

[Ali @amohammadpoor]

This has to increase conversion by double digits

[Jaytel @Jaytel]

I built myself a chrome extension called Pose

Any clothing model of any store becomes me.

Each brand still conveys their brand aesthetic, but I can quickly understand how something would look on me.

[Jaytel @Jaytel]

What started as building a personal taste.md skill for myself, turned into building a pipeline to create any taste as a skill.

The most important piece is references. This is where you should spend time. If the references suck, so does the skill.

I find that references cropped tightly on details in high resolution work the best.

Each image gets analyzed by both Opus 4.7 and GPT 5.5. The analysis is based on why the reference is successful as a piece of design - not what it does functionally. Using two models helps rule out biases and gaps from each.

The models focus on layout, spacing, typography, rhythm, composition, hierarchy, etc.

At the end, each image has:
reference-01/
- opus-4-7-analysis.md
- gpt-5-5-analysis.md

Then we fuse them together using GPT 5.5 - but the md files are anonymized so 5.5 doesn't prefer itself.

reference-01/
- fused-analysis.md
reference-02/
- fused-analysis.md

etc.

After fusion, we have one synthesized analysis per reference. Now the goal is to combine all of those into a single rule set.

This is where chunking matters. If you ask one model to combine 100 image analyses at once, the result becomes too broad. It summarizes instead of preserving the granular design rules we want.

Instead we chunk the fused analyses into smaller groups. Each group gets merged into a chunk-level synthesis, usually from around 6 to 8 image notes at a time.

Then one final model pass fuses those chunks into a single md rule set.

Finally, using the rule set, we write a skill of concrete instructions. It enforces constraints, uses imperative wording, and avoids vague taste words.

[Alim @almmaasoglu]

@tenobrus Persian islamic artisans were onto something centuries ago

[Brotzky @brotzky]

Introducing https://t.co/ycxJEf1z7w!

A new space where I explore how the best apps in the world are built.

First piece:
How's Linear is so fast? a technical breakdown.

https://t.co/9Vu1syrn1i

[Ali @amohammadpoor]

@thinkwithmark EUROPE WHEEEEEEN

[Mark™ @thinkwithmark]

We raised $1M dollars to reinvent how people read. Introducing Mark II - a $159 AI bookmark. Thread below

[Ali @amohammadpoor]

@amorriscode @bcherny Perfect, appreciate you responding!

[Ali @amohammadpoor]

Annoying error that breaks workflow, especially when designing in Figma and using MCP. No message goes through. /compact works as an escape.

Any fix for this? Maybe a command to clear images/attachments? @amorriscode @bcherny