Dr Dom Pimenta MD

Most AI in healthcare is built by people who've never had a patient deteriorate on them at 3am. You can tell.

@reidhoffman The funnest if nothing else

The best way to test people in conversation is to ask them a question you're deeply thinking about too.

@RaminNasibov Fire was probably a biggie. Penicillin/vaccines. The internet AI

Which invention changed humanity the most?

@hthieblot Msn

Anyone who surfed the early web between 1995-2010. What’s the one website/app you still think about?

@mal_shaik Being helpful IS networking (and good soul food as well)

the best founders i know dont network they just help ppl and somehow end up knowing everyone turns out being useful is the best growth hack

@paulg I am “hiring” my first full time commercial agent this weekend (building). It’s amazing how dumb and how smart the frontier models are, in a very particular and unintuitive way. Increasingly if you haven’t walked that geography meaningfully you’ll never understand it.

The only thing worse than having the CEO knee-deep in building stuff with AI is not having the CEO knee-deep in building stuff with AI.

@garrytan @dessaigne The score takes care of itself.

"The hack to raise is to build a company so good that raising becomes easy." —@dessaigne

@alexanderrX_ Is there a bilingual translation American > British?

british understatement is unique and i love it. if you work with brits you absolutely have to understand the translation: "this is probably not ideal" -> this is on fire "it might be worth reconsidering" -> stop, do not ship this "i have some minor concerns" -> i have been thinking about nothing else for four days once you crack the code the feedback is actually some of the most precise communication you will encounter.

@Yuchenj_UW How though? How are they doing this?

Google is fighting every final boss at once: OpenAI & Anthropic in models, Nvidia in chips, AWS & Microsoft in cloud, Meta in ads, Tesla in self-driving, Apple in phones and OS. At $4.6T, it feels weirdly undervalued.

@hthieblot I think industry is important here - regulated med device/pharma tech - 3x multiplier difficulty, vs legal tech (2x) or fintech (1.5x) or ad tech (0.7x)

Difficulty Levels of Startups: Insane: Robotics Hard: Hardware Medium: Consumer Software Easy: B2B SaaS

Don't forget to chill out this weekend.

@edwardlando Click bait 😂

Like many VCs, I really do believe Anthropic and OpenAI are the last startups. There is no point to building anything else. Everything has been solved or will be shortly.

If people don’t actively tell you you are being too hard on yourself, you are probably not being hard enough.

I love a crisis. Some people do, some people don’t. Hard to explain why, but IYKYK. 😂

@garrytan The game is as it ever was - solve the problem.

Founders must stop trying to building 2010-era businesses with 2026-era technology. Don't try to rebuild Foursquare or Yelp. Don't try to recreate Basecamp by 37 Signals with $10/mo SaaS pricing. Don't underprice! If it works it's worth a lot more. Don't be tempted to become "Tech enabled PE" with revenue tricks. The rules of tech changed with AI. Play the new game.

The harder it gets, the bigger the challenge. The bigger the challenge, the funner the game. We all played S Level video games for a reason 😂

@bettercallsalva I can 99% guarantee you that this problem will be solved by better token use and more thoughtful planning - and probably by better compute- within 12 months

@ericries Pre ordered!

A true OG! That is a true collector's item, I'm not sure I even have a copy anymore.

Most of optimising for investment does the opposite - if your business is on fire, it needs your attention, not investment. -> and if your business is ON FIRE 🔥 then investors will come to you. So either way it's probably a waste of your time.

@mcuban Yup - super value based care built foundationally on AI is coming though, it is inevitable. We are building it. Would love to chat with you more.

Right now, on the business side , AI is being used to find more ways to extract more money from every stakeholder they can.

Ernest Codman (1869–1940) was an American surgeon and early healthcare quality pioneer best known for creating the “End Result Idea,” which argued that hospitals and doctors should track patient outcomes and analyze errors to improve care. Despite this being written over 100 years ago, we fail to do this in most systems today. Despite decades of digitization, joined-up systems discussion, and integration, we haven't created a culture of learning from our mistakes as clinicians - in fact, quite the reverse. Whenever there is an SI, every single person in that chain of care should be notified so we can all learn. In the AI-first future we are moving towards, perhaps we will finally have Kaiser's 'Learning Healthcare System' we have talked about for a century.