Jon Tirsen (@tirsen) header
Jon Tirsen (@tirsen) profile picture

Tech lead of @blocks. Previously at ThoughtWorks, Google Wave and founder of @triposo. Love building stuff that users love.

Stockholm, Sweden
Joined January 2007
I'll be there
Come try Builderlab with the team 📷Nov 10, San Francisco The stack so far: - @jack on intelligence - @bradleyaxen on how Builderlab works in practice - explore and experiment with Builderlab live Join us for a hands-on afternoon at the Builderlab Studio: builderlabstudio.rsvpify.com…
21
markdown with frontmatter is the .exe of this era
3
100
Life hack: Dissolve creatine in hot water instead of cold. No more drinking sand! (And no, it takes hours to devolve inte creatinine even in boiling hot water.)
2
146
one of the most delightful user experiences I've had in a long time. taste is in fact a moat.
18
26
4
631
33,875
Jon Tirsen retweeted
I’m so fired up about Cash App tags. It is so incredibly, uniquely Cash App. The first time I used my wand to pay for something was at a 7-11. The very bored clerk rang up my order and when I took the wand out and tapped to pay he smiled, laughed out loud and asked what is that?!? That response has been near universal. Doesn’t matter the age, gender, or background. People light up when they see it. This is a really cool technical accomplishment by our hardware team. It’s not trivial to build such a dynamic piece of hardware. My daughter is obsessed with it, adults who see it ask what company makes it and how it works. It just resonates. There are so many possibilities to explore here. Another revolution in payments technology fueled by the amazing hardware, design, engineering, and product talent at Block. We just keep shipping.
pay by magic
10
13
2
199
30,061
This is pretty neat. cash.app/tags
2
128
What do you call it when you DON'T use an AI to generate an image? Or no one does this anymore?
78
By that argument humans aren't reasoning either. We're just reproducing things we've experienced previously in our lives. I think the main difference between "the best possible LLM" and the human brain is that our neural networks evolve over time. An LLM only has the prompt.
This quoted post is unavailable.
515
Hello Sydney!
1
336
LLMs are really bad at saying “I don’t know”. Instead they make shit up that sounds plausible. Early in my career I used to be exactly the same but these days I say “I don’t know” all the time. Glad that I can’t quite yet be replaced by an LLM.
3
3
23
1,683
I am still surprised by the mindblowingly amazing super powers of just listening to people.
2
22
2,498
He might be batshit crazy but at least he has good taste in döner. (My office used to be two blocks away.) exberliner.com/english-news-…
322
It's now possible to block ChatGPT's crawler on any website you control. Added the block: it's a no-brainer. Why? ChatGPT cites no sources. It's a one-way relationship where OpenAI takes what is published on the internet, and then doesn't give a single reference or link back.
104
1,266
108
4,532
716,625
Wait the rocket engines and the explosions make sound in space in Starfield? Pfft I thought this was hard scifi.
1
690
I've been using ChatGPT more lately and it's pretty clear why Google should be absolutely terrified.
1
2
786
And finally ChatGPT:
2
547
Google was working on this when I was there more than 10 years ago. How could they have fallen so far behind so quickly?
1
1
297