Started FAQify on 31 Oct. It went live on 1 Nov.
It's time to upgrade your website with FAQify. Keep your FAQ. FAQify covers the questions it misses.
When a visitor asks one, it answers in 1-3 sentences from your own pages and links the page it used.
Then it shows 1-2 real customer testimonials that back the answer. The AI never writes or rewords them.
Your dashboard lists the questions visitors asked and marks the ones your site doesn't answer.
Started FAQify on 31 Oct. It went live on 1 Nov.
It's time to upgrade your website with FAQify. Keep your FAQ. FAQify covers the questions it misses.
When a visitor asks one, it answers in 1-3 sentences from your own pages and links the page it used.
Then it shows 1-2 real customer testimonials that back the answer. The AI never writes or rewords them.
Your dashboard lists the questions visitors asked and marks the ones your site doesn't answer.
claude-mem got 1139 GitHub stars in a day.
It records what Claude Code does in a chat and feeds a summary to the next.
I uninstalled it one week after using it. The summaries used a lot of tokens in the background.
Now I keep lessons in CLAUDE.md. When I threw out my agents' first front end, I added one that day: no screen gets built until I pick from 3-4 mockups.
A new chat reads the file and picks up where the last one stopped.
Installing a skill doesn't mean your Claude Code agent will use it.
I ran 4 agents in parallel to build an MVP and told each one to load every relevant skill. The front-end agent didn't use the frontend-design skill.
Now I name the exact skills in each agent's instructions. For my next front-end build that was 11, frontend-design first.
Each agent loads them before it reads any code and writes them at the top of its notes.
My main Claude Code session checks each agent's terminal for those skills before it accepts the work.
5 Claude defaults I keep off my website when designing frontend.
Banned in my CLAUDE.md:
- off-white with a violet accent
- customer avatars that are just initials
- pulsing "Live" badges
- fake dashboards
- em dashes
I wrote the list the day I threw away my first front end.
Computer-use agents shouldn't need screenshots.
Vercel Labs' agent-desktop drives desktop apps through the OS accessibility tree, not pixels.
If a machine can't read your app, an agent can't use it. My AI Visibility Checker tests the same for websites: 11 AI bots.
@averycode check robots.txt first. GPTBot is OpenAI's training crawler. The ones AI search uses to fetch pages are OAI-SearchBot, Claude-SearchBot and PerplexityBot, so allow those. and since it started as an app, turn JavaScript off and see if the text still shows up
@guillemcraft I'd leave the close-button offer untouched while you change the rest of the paywall. It's 1 in 6 of your paying users, and if you change both at once, you can't tell what did what
Honestly, it's a real challenge for me too.
Trying to keep it genuine. I share what I'm building and the mistakes I make along the way, so people can get started faster and skip a few of them.
Latest one: FAQify, https://t.co/GzcOTfiexX shipped in 2 days.
Would love to connect 😄
I studied mechanical engineering, then analytics at Duke. Then I fell in love with building things.
Since 26 Oct, I've shipped 2 AI products with Claude Code:
AI Visibility Checker tells you if ChatGPT and Claude can read your site. Live in a day.
FAQify answers the questions your FAQ misses, from your own pages. MVP in 24 hours, live in 2 days.
I believe in shipping fast. You find out quickly if you have a winning product, and if you don't, you haven't wasted your time.
I post how I build: the agents, the prompts, and what broke.
Your FAQ answers 8 questions. Visitors ask the 9th.
I built FAQify for that. For questions your FAQ misses, it answers from your own pages, links the source, and shows a customer who said it.
Try it: https://t.co/sTPBKTRKLX
The first 10 founders get a FREE month! DM for code.
Most AI video tools generate from a prompt. This one reads your recording instead: your actions, your speech, and the room around you.
This will definitely redefine AR.
Introducing Reality-Aware Rendering in @Simulon, a completely new way to create with the real world.
Place neural assets with your phone and act out the scene. When you stop recording, Simulon renders the take, reading everything in it (what you do, what you say and the place around you), and your assets respond to all of it.
The final is regenerated from your recording: a new performance of the scene, not a copy of the footage.
Available now on iPhone. Coming to Android soon.
Introducing Reality-Aware Rendering in @Simulon, a completely new way to create with the real world.
Place neural assets with your phone and act out the scene. When you stop recording, Simulon renders the take, reading everything in it (what you do, what you say and the place around you), and your assets respond to all of it.
The final is regenerated from your recording: a new performance of the scene, not a copy of the footage.
Available now on iPhone. Coming to Android soon.