Dhruv Sumathi

Jensen Huang just tried our self-driving golf cart! thanks Jensen! @nvidia

we got @karpathy to drive our self-driving golf cart!

Working on this with @FynnComerford has been one of my favorite life experiences. Science is filled with stories and emotions that weave their way in and out of the work itself. We wanted to communicate some of these.

Letting a provider see all your data is the price of admission for AI. We're changing that. Introducing Silo, the first private post-training and inference stack for frontier models, with hardware-level guarantees that we can’t see your data. Privacy without compromises. 🧵

Introducing Trellis for Kimi K2 Thinking. It's post-training code that's 50x faster than the best single-node open-source version and 2x cheaper than training APIs. After safety testing, we're open-sourcing it, giving builders the best tools to customize a frontier model. 🧵

So much respect.

Built a tool that lets you convert a dense paper into a lesson written in the style of Richard Feynman! Supports highlights, notes, and inline deep dives. feynmanize.com

Wait a minute. The fact that we have a subconscious means either: a) certain patterns of brain activity (like the kind happening in our subconscious) do not possess consciousness or b) they do possess consciousness, but are a separate consciousness that we don't have access to. Another conscious entity, in your brain, that you can communicate with, but that isn't part of the consciousness that you feel. Am I missing something?

Really awesome essay by @nivitha22 about extreme heat and farming in Imperial Valley. Cool when people just go somewhere to observe how something works IRL. Feels like a new kind of "you can just do things" journalism. "Imperial Valley’s economy, from its $2.8 billion agricultural output to its emerging lithium and energy industries, depends on human bodies operating in extreme heat. Yet, the same people most critical to California’s food system are often the least protected from the climate hazards that now define daily life in the Valley." substack.com/home/post/p-18…

to the product designers out there: we can talk to sand but opening one of these currently requires either a sharp object or a blood sacrifice

In "The Technology of Liberalism" (lnk at end), I argue we should differentially advance tech that promotes liberalism (something like: everyone have their own inviolable sphere within which they are free) - especially as tech makes it easier to violate boundaries & centralize

Wild that there is a whole part of your brain’s activity (your subconscious) that (surprise) isn’t part of your conscious experience. Fun question: if there were other conscious experiences being produced by your brain, separate from the main one you feel, would you know?

@FynnComerford thanks 🥰

« The Earth exhaled and a thousand silver sails unfurled into the infinite sunlight, fanning outward past new galaxies, physical laws, beings terrible and magnificent. Around them worlds formed and fell, bloomed and vanished. And through their titanium irises the universe bled like starlight through glass. » wonderful short story by @Dhruv_Sumathi

@nivitha22 But can Dietz and Watson solve my SOC2 compliance issues

First short story! substacktools.com/sharex/SZZOibVL

@kanjun Congrats guys!! So excited to try this

Sculptor: the missing UI for Claude Code 🎨 Imagine running 5 Claudes in parallel, safely in containers, while you stay in flow. Then bring their work straight into your IDE to test/edit together. This is how one developer ships like a team. Try it with Sonnet 4.5!

FAF AI hackathon in SF! Meet other cracked builders from Stanford, MIT, etc. We’ll fly you out if you’re not in SF. $20k pool of cash prizes. Hosted at Delve’s new HQ in SF, with people from Cursor, Cognition, Felicis, a16z, Founders Fund. Apply here: hackathon.friendsandfam.xyz

With @luke_drago_, I’m cofounding Workshop Labs, a public benefit corporation preventing human disempowerment from AI. See below for: -impact case -what we’re building -what we hope the future looks like -what we’re hiring for

Co-hosting a hackathon at @EdgeEsmeralda on solutions to the Intelligence Curse — join us to think critically about changing the default AGI future!