Agent taming prompt whisperer. 🗣️💨🤖 Fancy seeing you here! Since you're here, do me a favor and hit that Follow button. ⬇️ It's free. 😎
South Florida
Joined February 2011
- Tweets723
- Following257
- Followers127
- Likes635
LLMs operate outside of time. To an AI, everything is an instance. Time only becomes a factor when concepts like memory are involved. Therefore, a prompt doesn't need to be in chronological order. After all, we are working with non-human intelligence. This reminds me of Arrival.
I'm thrilled to announce that @speechischeap ranks #1 on @ultrametricai! My favorite part about Ultrametric is it provides clear criteria of what the product is being evaluated on. And addressing it becomes a win-win-win for the consumer, the product, and the ecosystem at large.
Exponential growth in AI innovation combined with an ongoing global shortage of GPUs, RAM, and SSDs is the forcing function that's needed to make even ten year old hardware do incredible things.
I recall being excited about this idea 13 years ago. Thankfully the patent has long expired. patents.google.com/patent/WO…
I hate using multiple smart device apps to control the smart devices in my room, so I setup a raspberry pi server, reverse engineered a couple of protocols, added some UX magic, designed an AR enabled app that lets me go “lumos” and turn the lights on
So far I have integrated: Smart bulbs, Smart strips, AC, Bluetooth speaker, air purifier, and a couple of lamps
TIL instead of clicking my website's CTA, visitors were clicking on the dollar amount, which wasn't getting them anywhere. The lesson here is, if you don't have a heatmap on your website, you're flying blind.
This is exactly what I was expecting. Kudos @t3dotcodes!
Source: deepwiki.com/search/if-i-run…
Jules was a great way to tokenmaxx Gemini. I used it a lot before switching to Opus 4.8+ and later GPT-5.5+ in Q2. It's a shame to see it join the @killedbygoogle graveyard, but I'm not surprised in the slightest.
When we launched Jules, it was an experiment to learn how asynchronous coding agents can help developers ship faster. Seeing what you built taught us so much about where agentic development is headed, directly shaping several products across @Google.
We’re working on bringing our learnings from Jules into @antigravity so you get the best of both worlds: autonomous remote background task execution and an agent-first development environment.
As we focus on this unified experience, Jules will be winding down.
Whether you used Jules to crush bugs, write tests, or launch features— thank you for shaping the future of agentic coding! 💙🦑
My main keyboard looks similar
mapsdata.ai was never meant to be a saas.
It’s a tool that me and @KeshWelch built to pull leads internally.
Then I let some people try it and realized the demand.
Now we’re at over 1000 users and have pulled 10,000,000s of leads.
Super cool to see this software grow.
Much more to come.
I'm BLOWN AWAY we can do this sort of thing now!
Interactive 3D render: cdn.ilyakam.com/perm/toothbr…
Full chat history: chatgpt.com/share/6ae21c11-5…
IMO Claude Code should be renamed to Claude Do to avoid having to explain it's not just for coders every time. #ClaudeCode Inspired by @ben_vs92
After a four-month hiatus, I came back to check out @claudeai. Looks like I can't even quote another agent just to have a sidebar conversation about it…
It worked after making a slight modification.
Those bouncers are truly insane! Context: teapaw.org/theo/status/2078223917…
FABLE IS THE SAME MODEL AS MYTHOS.
I get this question a ton and I've even seen other "big ai influencer" types spreading misinformation about this.
Think of Fable 5 and Mythos 5 as two separate entrances to the same place (confusingly, also named Mythos 5).
When using Fable, Anthropic has more safeguards in place. These guards are NOT part of the model. They are layers in front that check your request to make sure it's safe. They also monitor what the model is doing while it responds to make sure the response is safe as well.
Mythos 5 has way fewer checks. It is a special entrance for trusted parties who have been given access with fewer restrictions.
That's the only difference. They are the same model. Same weights. Same servers. Same everything.
The difference is at the entrance.
Here are the unicode symbols I use to keep track of my chats in apps like @t3dotcodes, @ChatGPT, @claudeai, @GeminiApp, @perplexity_ai, etc.:
» - active task (pinned = higher priority)
✓ - complete
⑂ - fork from another thread
× - dead end
⧖ - waiting (plus the culprit)
Woke up to a tweet about an automated startup. Got my SaaS ranked #5 in a brand new category among peers later that day. Thanks @ultrametricai and the autostartup is a neat idea, @judegomila!
Your agentic chat conversations AKA "traces" are worth a lot of money. Take a minute to make sure you're not giving them up for free if they're backed up on GitHub.
Here's how:
1. Go to github.com/settings/copilot/…
2. Disallow GitHub to use your data for training.
