Joined May 2014
- Tweets 10,774
- Following 539
- Followers 479
- Likes 29,674
1,804 Photos and videos
Pinned Tweet
Apr 22
2
108
Jun 13
After a few days of prompting composer 2.5 on @cursor_ai I have the start of something dope for assetto corsa
1
40
Jun 13
I'm so stoked on this.
One step closer to finding out "can I drift this road"
11
Jun 13
LLMs are bell curve devs. We will soon find out how much the right quadrant carries software 👀
Jun 13
It's genuinely incredible how bad these models are at programming
1
120
Jun 13
Anthropic: mythos too dangerous. We should all slow down ai research for security.
Also anthropic: the gov't made me turn it off
Jun 13
The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.
The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.
Access to all other Claude models is not affected.
We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.
Read our full statement: anthropic.com/news/fable-myt…
1
36
Jun 12
Real ones will watch this and go hell yeah
15
Jun 12
Social media is like a radio.
As you turn the dial, it's mostly noise.
Sometimes you tune into a signal that you're not interested in.
Then, you finally tune into the signal you want and it's a vibe.
Dont listen to noise. Find the signal that fits your vibe 🤙
1
1
19
Jun 12
Welp we all know what this means
Jun 11
BREAKING: Jim Cramer says that SpaceX could soar to unsustainable levels after its debut
1
20
Jun 12
This is indeed interesting
1
1
7
437
Jun 10
Ok I'm bullish on google again. They are looking ahead.
- LLMs are not the final model that will give us better intellegence
- Openai and anthorpic are just throwing more compute at their models not innovating.
- Cloud models are expensive. That's okay for enterprise but not consomers.
Meet DiffusionGemma âš¡ Our latest experimental open model (Apache 2.0) that generates text up to 4x faster.
Instead of predicting and typing just one word at a time like most language models, it drafts and refines entire blocks of text simultaneously.
Here’s how it works 🧵 ↓
2
42
Jun 10
Not worthy
Jun 10
Our Anthropic overlords deciding which prompts the peasants are allowed to use.
16
Jun 10
Fable: short, fictitious story designed to teach a practical or moral lesson.
There's definitely a lesson to be learned in the "hallucinations"
219
Jun 10
The ole fable of getting left behind.
1
45
Jun 10
Just found a regression in some AI code.
User reported a bug so of course I put an agent on it.
Agent says nope it's all good no regression.
Old: UserPhoneNumber.where(num: param).last
New: UserPhoneNumber.find_by(num: param)
Can you find the issue?
13
Jun 9
You should be ruthlessly eliminating bad patterns and tech debt so you're loops doing multiple those patterns exponentially.
LLMs don't think. They regergitate.
Here’s your monthly reminder that you shouldn’t be prompting coding agents anymore.
You should be designing loops that prompt your agents.
1
1
42
Jun 8
When SpaceX drops the first orbital data center we will no longer deploy to "the cloud" we will deploy to "the stars"
20
Jun 8
Who cares if AGI is soon as long as the next episode of the office ai wars edition drops
Jun 7
Michael Scott, the most qualified person to judge Claude vs Codex, has entered the chat.
1
26
Jun 7
Indie hackers what's your stack and deployment to minimize server costs until profitable. Performance and background jobs required
28