HN user

yaseer

1,880 karma

Co-founder axiom.ai, YC W21 https://axiom.ai/

Outside of work, I'm extremely interested in automated theorem proving, the lambda calculus and its applications to computational biology.

An out-of-date blog explaining my old research is here: https://yaseersheriff.github.io/

Posts10
Comments391
View on HN

It says "Our users are everywhere" and shows some logos for the companies these users are from.

If the users are from those companies, this is not lying.

If they added logos for companies their users are not from, it would be lying.

Adding a logo to your webpage has started to follow different patterns for the stage of the company.

Early stage companies show things like "people at X, Y, Z use our product!" (showing logos without permission), whilst later stage ones tend to show logos after asking for permission, and with more formal case studies.

They may not have asked for permission to show these logos, but that's not the same thing as lying.

I've found writing docs and updating docs a great AI use-case.

In my experience documentation generation has a lower error rate than code generation, and the costs of errors are lower too.

I'm not really a big fan of AI agents writing features end-to-end, but I can definitely see them updating documentation alongside pull requests.

Super interesting, thank you very much for sharing your thoughts!

HN is still one of the few places on the internet to get such esoteric, expert and intellectually stimulating content. It's like an Island where the spirit of 'the old internet' still lives on.

It's not quite like other infectious diseases (e.g. COVID), in that transmission is dependent on mosquitos as a vector.

If they've sufficiently damaged the vector one tourist alone cannot bring it back - the disease vector would also need to come back.

I think that was written as a generic, illustrative example. That Example was also for healthtech, which OP is not.

Yes ex-Meta could imply non-technical, but OP is a CTO with a Stanford computer science degree. Their job title and skills are technical.

Am I missing something, or are HN readers assuming (incorrectly) the author is non-technical and looking for someone to build out their ideas?

From what I can see, the author is a CTO who is ex-meta and Stanford computer science. They look like a technical co-founder to me.

I know there's this annoying and widely prevalent persona of co-founder dater with only ideas, and no skills to build them, but this author doesn't seem to be that.

As a long time HN reader, eventually participating in YCW21, the main thing i would like to highlight is there's so many misconceptions about YC on HN.

It would take too long to refute them all, but just remember the HN readerships' view of YC is pretty distorted. ("it's all about VC money" , "it's just for ivy league grads" etc).

I could go on, but just read Paul Graham's essays - that's probably the best way to understand how YC thinks.

Unfortunately the advertising was so awful I just navigated away.

This site was close to spam on a mobile device.

If anybody has migrated from AWS to Hetzner what was your experience? What kind of cost savings did you see?

I know it's not a like-for-like comparison, I am particularly curious about the price differentials though, AWS is often a premium.

Having been through both Techstars and YC, this article is painfully accurate.

Being part of both networks has been an interesting experience in culture.

YC culture fosters an extremely strong network of founders, even outside the bay area. In London we're constantly pinging each other for advice - it really helps.

Techstars does not have this and makes little effort to do this. I remember speaking to some Techstars people in London trying to see if they might Foster more of the community here, but they had no interest in it unfortunately.

It's a real shame because some Techstars programs with the right MDs had YC quality (Shoutout to Connor), it's just the quality wasn't distributed or cultivated uniformly. It's heavily dependent on the MD of the program.

This is pretty interesting.

I play a lot of games, but never got into fighting games.

Seeing the strategy "laid out" might be a useful way to understand the mechanics rather than figuring them out with trial and error.

Trial an error can be fun (e.g Zelda), but it's always been something that stopped me getting too far with this genre.

This doesn't seem unreasonable for a limited supply product, to prevent price scalping.

If you saw these terms for a console launch or graphics card launch, you wouldn't complain.

It's probably not enforceable, but the idea isn't crazy.

EDIT: I was referring to the resale clause. Just saw the remote deactivation point too, which really is awfully dystopian.

Given the shitty Orwellian potential of technology though, I wouldn't be surprised we end up steaming OSes from cloud on mobile dumb terminals. But that's shitty.

I've been doing this for years with my tablet. The tablet honestly becomes 'less dumb' in the process though.

I agree some stuff moves towards this Orwellian centralisation of power - I use Geforce Now for games + Amazon Workspaces for work.

But I've also run VNC on servers I own myself for side-projects- you can own the dumb client and smart server if you need to.

When you setup your editor and container system to work on a remote dev environment, I've generally found this superior to coding locally - with a few edge-case exceptions.

You can't code when offline. But I generally find it difficult to code without internet access these days anyway, and the scenarios without internet access are now vanishingly small. Internet access whilst flying, is becoming standard, and only likely to improve.

The advantages are numerous. Firstly, the ability to work from even the most basic machine, anywhere with internet, is liberating. I can be perfectly productive with only my tablet whilst travelling.

Secondly, keeping a consistent and controlled environment that you never have to maintain or worry about breaking due to a system update, or move to new hardware.

If you use multiple devices(e.g. Desktop, Laptop, Tablet), or perhaps have personal and business machines, not having to repeat a setup and update process is extremely convenient. As is not having to worry about your hardware - I can code from any machine capable of running a text-editor, SSH and a web-browser.

The key thing here is the development flow needs to be extremely convenient for the developer - no frustrating processes interrupting your code-debug loops.

But when it's executed with convenience in mind, I've found it superior in every case, and in a future filled with ubiquitous internet, I wouldn't be surprised if this becomes the norm.

I'm wondering if PHP's reputation has been rehabilitated somewhat in recent years. At least amongst those who understand the significant changes in modern PHP, and the difference a stable framework like Laravel makes.

I've found PHP with Laravel a stable and solid workhorse. Good for getting things done quickly, and then a stable system over time.

In contrast, I've gradually become disillusioned with the node.js and JavaScript/TypeScript ecosystem over the years. You can build quickly, but the systems turns into a pain to maintain and update.

At first I thought "This sounds stupid, just play a little, have some fun, and move on like everyone else". Then I remembered my own experiences with addictive substances.

Addiction rewires your brain, which is an experience I forget when I am not addicted to something.

It's easier to not indulge at all, than try to have a little and attempt to control it from there.