keeptrying

My deepest condolences.

Much strength to you.

Yes the clinical trial system is super opaque and it's not clear what benefit one can get if one doesn't understand the process or even the risk profile of it.

This lack of patient agency applies to all of medicine, really.

I plan to improve patient-doctor relationship in general. Gut tells me that enabling educative access to the basic sciences component of medical school and paring it down to patient specific focus, we can enable much higher patient agency through better patient-doctor communication and a deeper understanding of ones condition.

More importantly a better decision making process and ability to query and understand the doctor.

I've been through this with my wife and it's just so important.

Thank you for your writing - its taught me a lot about a lot of things. One concrete highlight is how important patient agency is in the patient-doctor relationship - which you've written about a few times.

I'm truly deeply sorry about this whole situation. Thank you for sharing all your knowledge.

This paper is super interesting and seems to agree with real life anecdata - most managers are crap and management is a hard skill that people are not trained for.

How to find a good engineering manager to work for:

1. during your interview you must learn something significant - can be technical, or about managing people or something.

2. they should not grate you even 1 second. The interview is where you and they are at your respective very very best behaviors. If they grate on you even for 1 second, you probably don't want to work for them. You should always come out feeling positive.

3. they should push you during your interview without you feeling bad about it. The job of a manager is to bring the best/most out of you. When you are on the job they will push you hard. Also they need to find the boundary of your abilities during your interview.

4. they must have better communication skills than you and you should look up to them. Everyone judges their boss and can't deal with a boss that is worse than them at their job.

Find a good manager to work for is very difficult. Because of the high stakes related to a job search - finding a good one will probably only happen by accident.

Treasure those bosses.

The worst thing that can happen (as it happened with me) is that your first boss is great at their job. Then every boss after that will be such a huge letdown. :)

Whoa - Can anyone explain why I'm getting downvoted?

That is really freaking cool. Creating a platform that's location independent really helps people with disabilities.

I do think LLMs by enabling people to work indepedently (in adiditon to remote) should add degrees of freedom that would help people with disabilties.

LLMs should enable this by easing creation of work contracts (ie allow more accomodations by changing contracts), automation of most of most business operations, ability to quickly get to "average level" on a lot of business concepts, automation of finances (eg: using Runway).

Hopefully the creative piece of building a business will no longer get drowned out by the operational pieces.

I refused to interview at Amazon for policies like this.

I work very hard but I would need to be compensated or given other perks.

LOL - seriously? Nice.

In NYC they do this all the time. They'd also nitpick the hell out of your work.

You'd feel like a complete loser after your review. Pretty hilarious when thinking back about it.

At Amazon you get paid when you create whole new businesses which you used to pitch to Bezos.

Incremental improvement was rewarded through the regular stock and pay process.

Thus no one cared enough to quickly switch to LLMs.

Suoer interesting how org design - even when brillaint can be severely lacking.

True. WFH was the real trial of offshoring especially to similar time zones.

Also it took the risk off the CEO plate that remote might fail. Further the market is rewarding them for it now.

$1-$2M after 6 years of working at $100k isn’t really much either in the Bay. (Which is the only place you’d get that.)

Even that averages to a senior Eng salary for the very very few founders who get there.

This has to be tempered by other realities - no social life - working 80 hours a week easily - risking personal finances - health problems - good chance of divorce / no deep relationships

Starting a company is no joke.

Very few get to series B/C.

Tech jobs are going to move out of the bay at the larger companies.

Companies are being rewarded for being efficient in the stock market and CEOs are taking heed.

Founder liquidity doesn’t make up for much in the average situation.

Making $400k after making $100k for 4 years doesn’t really change much.

It gets you upto junior engineer level.

The underestimated play is becoming a cofounder to a great CEO 2nd time founder.

I implemented a BSP based renderer too in around 1998- 2000.

Probably from that same paper.

It was/is such a cool fun little algorithm to implement.

We were building virtual reality worlds 24 years ago and people are still not using it.

How do you build Stripe payments with LLMs.

Only very shallow SaaS may be killed with LLMs but not 98% of em.

I think their training methodology was suspect.

Also I think there’s a bunch of important techniques that just are t shared.

I dont believe that to be true. Rates of restraunts are double what they used to be. That can't be coming from just costs.

Food has become very expensive. McDonalds is part of hte problem.

McDonalds is no longer affordable by low-income consumers.

They really need to lower food prices. It's ridiculous that they are charging $15 for a burger, fries and a drink.

The bill for eating out at a restraunt for 2 people was $60 ... now its $120 minimum.

On the face of it this seems to imply, moving slowly and deliberately.

However this is not always true.

Example: Detroit lions Vs SF in 2023 playoffs. Avoiding blundering means being adaptable to the situation at hand - not using instinct derived from previous battles.

Wow, incredible achievement to build Postgres replication.

Something so many people use. Inspiring.

(Note to self: never mention genitals in an obit!)

I hope I build something that’s used at this scale.

RIP Simon.

If you are gonna be doing stuff like this I’m damn excited for answer.ai!

It’ll be the first time we’ll have someone who knows AI create leverage to open source it.

Way to go!

Good article highlighting what most of us have probably felt about the AI market - ie incumbents don’t have Innovators Dilemma here.

It’s unfortunate that he doesn’t actually delve into new strategies.

Anyone know of any actually good articles on plausible AI startup strategies?

The problem with this kind of system, is it takes away the interesting parts of the job for employees.

I mean, who the hell wants to do a list of tasks created by an AI?

Sierra AI - Bret Taylor’s conversational platform came out of stealth.

Kind of underwhelming. Was expecting something more innovative.

Also a competitor to OpenAI which he’s on the board of.

Slightly weird to say the least.

I was eager to work on this because at some level you will definitely be able to apply Reinforcement Learning but I was worried that OpenAI would definitely jump into this.

And it seems that they are focussing on tasks :

https://www.theinformation.com/articles/openai-shifts-ai-bat...

With enterprise, you can easily go 1 year with literally nothing to show for it.

If the sector has one adverse event in it, you won’t have much even after 2 years.

Small Buisness - at least there’s a lot of them and the constant interaction itself will propel you forward.

Great stuff. Spoken language learning is definitely the weak point of companeis like Duolingo and is the part of the language learning that most people want in the first place.

That lag in the video is still a bit too long.

If you can get that down to nearly instantaneous, I think this would have a lot of appeal.

Damn need to go get one. Still have the series 3.

Yes its more about Venture funding than YC specifically.

Thsi is important for all founders to understand.

If founders go into YC knowing the large risk they are taking - that’s all good and well.