Filippo (@neysofu) header
Filippo (@neysofu) profile picture

Bits mangler. Cera

Concent of Saunt Tredegarh
Joined May 2016
Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)Photo by Filippo (@neysofu)
Filippo retweeted
Without getting into too much detail about the conversations I’ve had with my friends at Anthropic, it’s become obvious to me that their long-term ambition is to continue running these small-to-large experiments while scaling experiment throughput with AI agents. Now, researchers already use smaller models to test architectures and optimizers. However, what I’m hearing is that they want to do this on a much bigger scale. Think millions of smaller, cheaper experiments with models that have a billion parameters, with the most promising discoveries being validated at 10 billion, then 100 billion, and so on and so forth. The idea is to have autonomous research agents explore this enormous hypothesis space and identify which architectures and training methods scale best. Now, this probably sounds obvious to a lot of people already in the AI field. However, from what I’m hearing, there’s a lot of success happening right now, and they’re continuing to scale up this experimentation, which is quite exciting to hear.
can tell from the latest haiku :)
23
21
3
645
33,307
we married on the 11/11 and @calcom just hit $11.11M in ARR 🔥
133
13
3
1,465
95,381
Filippo retweeted
>engineer says an agent swarm secured his software >ask if it’s formal methods or 8 million lines of smoke tests >he doesn’t understand >pull out illustrated diagram explaining formal methods and smoke tests >he laughs and says “it's good security sir” >check >it smoke tests
12
5
1
249
10,250
Composability really is the name of the game. We need to build a culture and ecosystem of open specs, as opposed to monolithic verification efforts.
The bull case for formal verification is "Here is a reusable library of properties/specs you can import in your code" not "Here is a few million LOC of Lean for you to go over, but the model passes!"
2
6
2
45
2,853
Filippo retweeted
The language matters more with agents, not less. The compiler is the only reviewer that reads every line
3
15
1
73
3,004
Filippo retweeted
💭 unpopular opinion: type theories of Lean / Rocq are not complex and heavyweight enough. we need low-level, type-checked control over runtime *and* programs and proofs in the same lang maybe dependently typed lang with OxCaml's modes. even better: graded modal dependent types!
Lean, Rocq, etc. seem very complex and heavyweight for many applications but they serve as least common denominators, and now that AI can write in type theories, why not? Still, it feels like a smell and I sense something different is around the corner, not sure what though
23
17
1
194
9,485
The bull case for formal verification is "Here is a reusable library of properties/specs you can import in your code" not "Here is a few million LOC of Lean for you to go over, but the model passes!"
32
22
11
305
26,128
That's right! Audio spectrography goes in the transformer hole.
So Astra is able to identify sounds from mel spectrograms zero-shot. I don't think we've scratched the surface of what this model can do (and this is light reasoning btw)
21
146
11
3,479
111,697
Replying to @SeanSankerJr
Current programming languages were designed with human ergonomics in mind. Humans do not write code anymore. So we should throw away all old languages and design new ones, geared toward the machines to write code in.
13
6
2
73
5,377
Filippo retweeted
one single "career strategy" that worked out incredibly consistently for me in the last 15+ years has been being comfortable with low status decisions. the marshmallow thing. so much alpha there's some real power in constantly being the underdog. being layered, underestimated. you can stay on the offensive all the time. less to lose. much more fun. so many more friends too! all pretty logical but it's so funny how much sf obsess over bullshit status games. putting people on status gradients. oh where you work, who are your investors, where did you go to school, otherwise i won't waste my time talking to you... it's so shortsighted. life is easier and more successful if you're not trying to be the cool kid
26
20
3
636
67,625
A whole generation trained on Factorio just to be ready when robotic automation is hitting its inflection points. You are the chosen one.
21
8
2
176
12,599
Replying to @exec_sum
there are hackathons with similar budgets every month in SF
2
61
3,636
Filippo retweeted
The high five wasn't invented until 1977. Never kill yourself
3
2
13
1,558
Filippo retweeted
Y'all know that the RAM prices skyrocketing is the first sign of the paperclipping, right?
208
316
74
20,198
3,288,595
Introducing Sovereign SDK -- the first rollup framework that executes 99% of transactions under 10 ms, fast enough to bring price discovery fully onchain. Now even solo devs can build efficient, global financial apps. Production-ready today. Go build the next Hyperliquid.
260
109
81
794
735,175
Filippo retweeted
Replying to @mitchellh
It’s just a couple of texture samples, Mitchell. What could it cost, 10 cores?
1
1
10
732
The fact that AI margins are much worse than software margins hasn’t been internalized by most companies
55
69
9
1,527
105,360
Cruft breeds cruft. (If you're not familiar with the word "cruft," it means an accumulation of unnecessary stuff.)
66
29
6
506
67,661
You are sheltering non USB-C devices in your carry-on, are you not?
It’s time for THE charger. Today, the USB-C becomes officially the common standard for charging electronic devices in the EU. It means better-charging technology, reduced e-waste, and less fuss to find the chargers you need. #DigitalEU

119
2,133
66
35,562
1,992,888
the fatal flaw in startup culture is the assumption that intensity can substitute for insight
7
19
1
156
6,157