one of the hardest parts of fundraising for founders is after they start getting a few no's they lose a lot of confidence in their business and the next set of investors can tell they don't believe.

understand why they passed but don't let them kill your belief.

Yep. Delta just totally fucked me. Prior Delta flight was delayed and then I sprinted across Atlanta airport to make it fifteen minutes before my flight and the gate agent wouldn’t let us board despite the flight not leaving the gate for another 20 minutes+

notion is one of the least intuitive pieces of software to gain traction in yrs

everything so god damn clunky because of the dumb underlying data structures they want to impose which should be completely abstracted from the user

truly awful and stunning we've let this catch on

a fundamental problem with all of digital health is that it is still feature incomplete relative to the traditional healthcare system both on a micro and macro scale.

most of healthcare is about how you handle edge cases and digital health is design to handle the 80%.

The approach from hardmaru and Sakana AI has a chance to be very disruptive to the chip-model-cloud provider industrial complex.

If you can combine lagging edge models with low compute to get SOTA performance on your task...

sakana.ai/evolutionary-m…

Worth a listen.

Mario Schlosser has a truly unique set of traits:
- deeply understands the healthcare system
- deeply understands AI
- has high emotional intelligence, humility, and is genuine
- has tremendous patience and resilience

I'm cackling at the realization I just had that Coca-Cola pays for a diabetes management solution for their employees.

If this were a science paper, you would expect a country that picks its science workforce at random as a “weak baseline” and a leading nation like the US to actively experiment towards state-of-the-art, or at least beat the baseline.

Not providing a guaranteed path for…

mediocre options are often the worst kind of options to have as you talk yourself into them

we're going to find there's a lot more eval rigging and testing on the training set going on than we thought.

beware what you see out there.

within 2-3 yrs max we will have an onslaught of scams. potentially as soon as the next 12 months.

talk to your family + parents. get in the habit of validating identify w/ a piece of niche information both parties know that is not on the web (e.g. some random story about you)

Very excited to be partnering with Tufts Medicine in a model that I think can be THE model for how health systems can work with startups to expand access to care broadly.

more than half of digital health is just people running around trying to opportunistically yeet products into the market

perhaps the least strategic/thought out market in the world

ok I don't want to pay 600 bucks for information pro who else of note is on the list besides everett, LM, and delian?

It feels like schooling should teach people how to reflect on when they're wrong and say sorry.

more than just change which everyone is now aware of but across the board the software infrastructure that runs healthcare is stunningly bad across the board.

0 incentive to change when the gatekeepers insist on owning equity in the software and have no incentive to fix it.

the scary part of this is an end state where we personalize the answers from AIs so much (bc consumers prefer AIs that agree with them) that the internet exists only to serve content that reinforces our existing viewpoints.

stratechery.com/2024/aggregato…