co-founder and forward-deployed cto @RepelloHQ | delusional optimist
San Francisco, CA
Joined June 2019
- Tweets3.3K
- Following1.5K
- Followers1.1K
- Likes61.4K
Naman Mishra retweeted
sam: local inference is a terrible allocation of scarce resources
“a lot of people want local inference, and somehow they assume that the problem is solved now because we have open-source models, we're going to be able to run them on our computers, and so we're not going to need all the stuff in the data centre
but from a global allocation perspective, actually it's quite terrible to have everybody, like, have a bunch of the world's limited supply of DRAM sitting in their own computer and being used five percent of the time
you want maximum utilisation in the cloud if you're trying to deal with this very supply-constrained environment
somehow many people's intuition is the opposite of that”
feels so humbling to know that i'm working on probably the single most important problem in this technology's adoption right now.
envisioned it 3 years ago, it's all coming together now
Naman Mishra retweeted
a lot of people ask for my $1M side project's tech stack.
it's pretty simple:
- MacBook
- Opus 5.5
- being totally fine with lying on the internet
you really only need that last one tbh
Naman Mishra retweeted
Can anyone help get access to Carbon as part of Fairwind?
@GoogleDeepMind
Just two weeks after rolling out Gemini 4 Argon, excited to share a preview of Gemini 4 Carbon - our next frontier model. It has state-of-the-art long horizon coding capabilities driving SOTA performance across the cybersecurity defense lifecycle including detection, validation, and remediation. It also ranks #1 overall on the recently updated v5.0 of Artificial Analysis Intelligence Index at 59.
We're starting by rolling out to a subset of cyber defenders through our Fairwind Program this week and soon expanding access further. Read more at: blog.google/innovation-and-a…
with so much happening since Oct 23rd in cybersecurity, one can say this has been one heck of a "cybersec awareness month" for sure!
we've been running at full throttle at @RepelloHQ closely observing, building, breaking and reporting stuff this past fortnight - a lot of exciting details coming soon.
meanwhile what should we name this cute mascot? open to suggestions!
P.S. : this was made entirely from claude-opus-5.5 by the very amazing @Arunkumar__23
Naman Mishra retweeted
If you were thinking concentrating AI behind closed doors is safe, it is time to update your priors!
EXCLUSIVE: In less than 12 hours, a group of ex-DOGE cyber researchers were able to penetrate Anthropic’s systems using AI agents — including Slack channels where the most sensitive models are discussed.
This is just weeks after
Anthropic whistleblower Joseph Coxon charged the company w/launching “a hubristic gamble” from a “private company’s Slack.”
Read our story here (gift link):
wapo.st/4smNlsC
Naman Mishra retweeted
The only thing that stops a bad guy with an abliterated LLM is a good guy with an abliterated LLM.
Naman Mishra retweeted
I don't think I'd have said it earlier but you really honeypots in your infrastructure as a bandaid to the huge agentic swarms coming at your company, all powered by abliterated OSS models.
Buy the time, but go fix it! Cybersecurity is existential to your business.
Naman Mishra retweeted
They really just got on stage and said this newly designed 4000+ pound car can do flips and go half the speed of sound (380+mph) and will ship in 2028. This is a prank, right?
