
Building @biscuit_so • Previously @Framer, @Spotify
AMS/BCN
Joined August 2007
- Tweets7.1K
- Following2.1K
- Followers3.8K
- Likes28.9K
We’ve been quietly building Biscuit for 8 months. Today we’re opening up the waitlist.
Biscuit lets you describe your business idea and we handle everything: product, payments, emails, hosting, going live. You can publish and get your first customer the same day. – at Barcelona, Spain
26x real time on an iphone. in safari. from a model that's under 40 MB brotli'd, with the same voices and quality as @GradiumAI's own runtime
every word is timestamped, so captions and read-alongs just work.
blixt.xyz/phonon/ (open source, apache-2.0)
phonon by @GradiumAI is just such a good local text-to-speech model. imo noticeably better than kokoro.
with int8 weights it's only 44 MB!
on a fresh page load with the model cached you can have speech in one of many voices in literally less than the blink of an eye!
26x real time on an iphone. in safari. from a model that's under 40 MB brotli'd, with the same voices and quality as @GradiumAI's own runtime
every word is timestamped, so captions and read-alongs just work.
blixt.xyz/phonon/ (open source, apache-2.0)
Blixt retweeted
You don't realise how good your product is until someone pushes its boundaries and shows a fire demo!🔥
Yesterday, Peter showed a demo that made us speechless. We were literally saying just - "wow", "oh", "whoa" 😄
He built multiple apps in @Biscuit_so but the standout app was the one where you could see a 3D model of a house with a terrain map for people planning to build new houses.
Blixt retweeted
Marom today built this to increase efficiency in choosing what to have for lunch. Ofc on @Biscuit_so
@blixt and @CompuIves we look awesome
join the input waitlist! i've been using input since early beta and you can really feel the artisanal touch behind every feature, interaction, pixel.
agreed, i caveat gpt feedback very often to opus ("was sol exaggerating", "please check real world uses", "could this happen in practice", ...)
Like a lot of you, I use Opus 5.5 to write code and Codex Astra to review it.
But Astra has a bad habit of flagging bugs that are technically possible but 𝗮𝗹𝗺𝗼𝘀𝘁 𝗻𝗲𝘃𝗲𝗿 𝗵𝗮𝗽𝗽𝗲𝗻 in practice. Opus sees the logic checks out and fixes it, then Astra flags the next one, and the loop never stops.
Add this line to your review prompt, skill, or AGENTS.md:
"Rate each bug from 1 to 10 on how often it would happen and how bad it would be. If the two scores multiplied are below 5, just file it as an issue and leave the code alone."
Tune the threshold of 5 to your project.
For Opus, the one fixing the code, add one more line to CLAUDE.md or AGENTS.md:
"Do not design for rare or future cases nobody asked about."
One of the better phishing attempts I’ve seen lately, watch out for this one.
Looks like legit calendly link
Redirects to plausible api-cal.com
Hides scary permissions below “can see who you follow” etc
Chain of failures here, from Calendly and X too!
Blixt retweeted
Replying to @Biscuit_so
@Biscuit_so team is in such a flow state that we have matching socks at panel events
@UvAFounderSpace thanks for hosting us
…2023: stable pre-ai job
2024: soul searching
2025: ai startup
2026: first 3/4ths look flat because the tail end is up to ~500 commits per day, imo a real inflection point
i wanted to visualize what my github history looked like before and after the AI inflection point.
see yours here - mo.rgan.co/gh/
i've looked at a few friends, eng and design, and it's really interesting to see the patterns.


























