Eddy Atkins

Machine Learning @sandstonehq

[next play @nextplayso]

(6/23) A few of our favorite under-the-radar roles from fast-growing companies:

• @AllenWangzian from @ditto_dates is hiring a head of marketing.
• @OfekLavian from @joinforage is hiring a consumer fintech PM.
• @CompleteSkeptic from @typesafeai is hiring a founding recruiter.
• @alypavs from @stripe is hiring a short-form video and social lead.
• @RSerna2010 from @finix is hiring a startups and VC partnerships manager.
• @businessbarista from @tenex_labs is hiring 50 people, especially AI strategists.
• @JayHoovy from @stanforcreators is hiring a General Manager and a chief of staff.
• @nifleisher from @sandstonehq is hiring salespeople and a forward-deployed engineer.
• @albrgr from @coeff_giving is hiring a senior operations associate to work in the office of the CEO.
• @AdamJJoseph from @clipbook_ai is hiring enterprise client success people and founding engineers.

✨ Follow Next Play for more (and link to full list in the first comment)

[Eddy Atkins @ecatkins]

@futurenomics If you have this much excess demand - aren't you pricing your round wrong?

[Eddy Atkins @ecatkins]

@joinshiftX shift

[BuccoCapital Bloke @buccocapital]

Lmao

[Eddy Atkins @ecatkins]

@tbpn's best superbowl ad isn't their "superbowl ad". This is brilliant

[TBPN @tbpn]

Introducing Claude with Ads

Free access to Anthropic’s most powerful model, powered by ads.

Try it now:
https://t.co/QU43cgIJRm

[Eddy Atkins @ecatkins]

@OpenAIDevs really cooked with the Codex app

🫡 @hellodzzy

[Eddy Atkins @ecatkins]

The easy wins are already priced in. We’re betting on the hard things.

[Liam Germain @liamail]

the first of many great pieces from the inner walls of the Sandstone AI team where we wake up every day with the intention of building something great for our customers

https://t.co/AopIC8P6Hu

[Eddy Atkins @ecatkins]

One of the more under discussed tradeoffs in building LLM driven systems (outside of code) is the performance-latency curve.

These models benefit so much from rich context, but I think users currently have expectations that make this difficult. Reset expectations or optimize?

[Eddy Atkins @ecatkins]

Having worked for both kinds of startup - the advantage in-person has over remote is incalculable - especially where there are hard, unobvious problems to be solved

[TBPN @tbpn]

"Remote is evil." - @martinshkreli

"If you're going to war together, you can't go remotely. And if we're going to take on Bloomberg, or whoever we're fighting, I just don't want to do it with some guy at home with his feet up."

[Eddy Atkins @ecatkins]

@tbpn @MartinShkreli @liamail

[Eddy Atkins @ecatkins]

Incredibly excited by our team @sandstonehq - 100% confident this is the group to execute on our vision

We will win by doing the hard things

[Sandstone @sandstonehq]

"Software has had its Claude Code moment.

In-house legal will have its Sandstone moment."
-@ecatkins, Head of ML

[Liam Germain @liamail]

Long time coming @tbpn @jordihays @johncoogan @RealProducerBen @nifleisher @cc_jarryd @sandstonehq @yaboiiyash @dev1_w @ecatkins

[Liam Germain @liamail]

LIVE FROM THE COURTHOUSE OF CODE, THE CHAMBERS OF COMPLIANCE, THE FOUNDATION OF IN-HOUSE LEGAL LAUNCHES ON @tbpn LIVE AT 4:30 EST

come see @nifleisher call our shot for what the future of in-house legal should look like!

@jordihays @johncoogan @RealProducerBen

@sandstonehq @cc_jarryd @sequoia @BogieBalkansky @rex_woodbury @daybreak_fund @mantisVC @AlexPallNY @TheChainsmokers

[Bogomil Balkansky @BogieBalkansky]

@nifleisher, @cc_jarryd and the ambitious team at @sandstonehq are transforming operations for mid-market in-house legal teams with an AI-powered control center. We @sequoia are proud to lead their seed round.

[Devon @dev1_w]

Excited to be joining Sandstone’s founding team—today we’re stepping out of stealth and announcing a $10M seed led by Sequoia.

I moved from legal practice to building legal software in 2020—still the "pre-ChatGPT" era of legal tech tools. Since then, generative AI has become increasingly powerful.

At @sandstonehq we’re building the AI-native operating platform for in-house legal teams—a context + infrastructure layer that unifies legal data, understands business context, and surfaces the right info exactly when you need it.

Huge credit to @nifleisher, @cc_jarryd, and @liamail for assembling an exceptional team of engineers, attorneys, and operators.

More soon. If you’re curious about what we’re building, let’s talk.

https://t.co/t7IbSEPqb4

https://t.co/fQcbmXYpas

[Business Insider @BusinessInsider]

While still at McKinsey, Nick Fleisher and Jarryd Strydom built a task tracker for lawyers and tested it with friends. It caught on. They pitched Sequoia on their last day at the firm. https://t.co/l0XxqgBDro

[Eddy Atkins @ecatkins]

@TheStalwart Bianca Censori.

Only because I went short this week

[Eddy Atkins @ecatkins]

@karpathy is a titan of clear thinking

[Andrej Karpathy @karpathy]

My pleasure to come on Dwarkesh last week, I thought the questions and conversation were really good.

I re-watched the pod just now too. First of all, yes I know, and I'm sorry that I speak so fast :). It's to my detriment because sometimes my speaking thread out-executes my thinking thread, so I think I botched a few explanations due to that, and sometimes I was also nervous that I'm going too much on a tangent or too deep into something relatively spurious. Anyway, a few notes/pointers:

AGI timelines. My comments on AGI timelines looks to be the most trending part of the early response. This is the "decade of agents" is a reference to this earlier tweet https://t.co/NiSn6jftqq Basically my AI timelines are about 5-10X pessimistic w.r.t. what you'll find in your neighborhood SF AI house party or on your twitter timeline, but still quite optimistic w.r.t. a rising tide of AI deniers and skeptics. The apparent conflict is not: imo we simultaneously 1) saw a huge amount of progress in recent years with LLMs while 2) there is still a lot of work remaining (grunt work, integration work, sensors and actuators to the physical world, societal work, safety and security work (jailbreaks, poisoning, etc.)) and also research to get done before we have an entity that you'd prefer to hire over a person for an arbitrary job in the world. I think that overall, 10 years should otherwise be a very bullish timeline for AGI, it's only in contrast to present hype that it doesn't feel that way.

Animals vs Ghosts. My earlier writeup on Sutton's podcast https://t.co/rSp1noyGBr . I am suspicious that there is a single simple algorithm you can let loose on the world and it learns everything from scratch. If someone builds such a thing, I will be wrong and it will be the most incredible breakthrough in AI. In my mind, animals are not an example of this at all - they are prepackaged with a ton of intelligence by evolution and the learning they do is quite minimal overall (example: Zebra at birth). Putting our engineering hats on, we're not going to redo evolution. But with LLMs we have stumbled by an alternative approach to "prepackage" a ton of intelligence in a neural network - not by evolution, but by predicting the next token over the internet. This approach leads to a different kind of entity in the intelligence space. Distinct from animals, more like ghosts or spirits. But we can (and should) make them more animal like over time and in some ways that's what a lot of frontier work is about.

On RL. I've critiqued RL a few times already, e.g. https://t.co/mYrMFVdVDW . First, you're "sucking supervision through a straw", so I think the signal/flop is very bad. RL is also very noisy because a completion might have lots of errors that might get encourages (if you happen to stumble to the right answer), and conversely brilliant insight tokens that might get discouraged (if you happen to screw up later). Process supervision and LLM judges have issues too. I think we'll see alternative learning paradigms. I am long "agentic interaction" but short "reinforcement learning" https://t.co/2L7FiaoKsw. I've seen a number of papers pop up recently that are imo barking up the right tree along the lines of what I called "system prompt learning" https://t.co/df5mJDdN3C , but I think there is also a gap between ideas on arxiv and actual, at scale implementation at an LLM frontier lab that works in a general way. I am overall quite optimistic that we'll see good progress on this dimension of remaining work quite soon, and e.g. I'd even say ChatGPT memory and so on are primordial deployed examples of new learning paradigms.

Cognitive core. My earlier post on "cognitive core": https://t.co/q2s1ihGy0T , the idea of stripping down LLMs, of making it harder for them to memorize, or actively stripping away their memory, to make them better at generalization. Otherwise they lean too hard on what they've memorized. Humans can't memorize so easily, which now looks more like a feature than a bug by contrast. Maybe the inability to memorize is a kind of regularization. Also my post from a while back on how the trend in model size is "backwards" and why "the models have to first get larger before they can get smaller" https://t.co/6k0FZRGXsb

Time travel to Yann LeCun 1989. This is the post that I did a very hasty/bad job of describing on the pod: https://t.co/fQgqaXPyp6 . Basically - how much could you improve Yann LeCun's results with the knowledge of 33 years of algorithmic progress? How constrained were the results by each of algorithms, data, and compute? Case study there of.

nanochat. My end-to-end implementation of the ChatGPT training/inference pipeline (the bare essentials) https://t.co/SIetgyoKWN

On LLM agents. My critique of the industry is more in overshooting the tooling w.r.t. present capability. I live in what I view as an intermediate world where I want to collaborate with LLMs and where our pros/cons are matched up. The industry lives in a future where fully autonomous entities collaborate in parallel to write all the code and humans are useless. For example, I don't want an Agent that goes off for 20 minutes and comes back with 1,000 lines of code. I certainly don't feel ready to supervise a team of 10 of them. I'd like to go in chunks that I can keep in my head, where an LLM explains the code that it is writing. I'd like it to prove to me that what it did is correct, I want it to pull the API docs and show me that it used things correctly. I want it to make fewer assumptions and ask/collaborate with me when not sure about something. I want to learn along the way and become better as a programmer, not just get served mountains of code that I'm told works. I just think the tools should be more realistic w.r.t. their capability and how they fit into the industry today, and I fear that if this isn't done well we might end up with mountains of slop accumulating across software, and an increase in vulnerabilities, security breaches and etc. https://t.co/8556ESSpyY

Job automation. How the radiologists are doing great https://t.co/FVUI872dkD and what jobs are more susceptible to automation and why.

Physics. Children should learn physics in early education not because they go on to do physics, but because it is the subject that best boots up a brain. Physicists are the intellectual embryonic stem cell https://t.co/p72Elk8lPV I have a longer post that has been half-written in my drafts for ~year, which I hope to finish soon.

Thanks again Dwarkesh for having me over!

[Eddy Atkins @ecatkins]

@DKThomp Small fact check. Thinking Machines launched it's first product literally yesterday

[Eddy Atkins @ecatkins]

@ModeledBehavior I would bet against it