frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

Open in hackernews

Some thoughts on my first YC Demo Day

https://billchambers.me/articles/yc-demo-day-spring-25/
33•kaycebasques•2h ago

Comments

mkagenius•2h ago
Why is this top post on hn within one minute of posting?
TZubiri•2h ago
Maybe karma affects post visibility?
mkagenius•2h ago
No, then he would keep posting whatever every day
sillyfluke•2h ago
I'm guessing the speed with which it received its initial few votes.

It's a threadbare article with little or no meat on the bone. So it is a little strange.

mkagenius•2h ago
It had literally two votes. Anyway looks like I shouldn't have brought this issue up, getting downvoted
sillyfluke•2h ago
Yeah, you're supposed to email hn instead of posting if you think it's an issue.

I've seen 3 point new posts up near the top before so I assumed that's what it was. 2 points top of the page is pretty aggressive tho, it might have been a near instant upvote by someone who was interested in the topic.

ge96•2h ago
Carl Weathers
Brajeshwar•1h ago
I don’t know how or why, but I have seen this happen many times. Many of my submissions appear at the front within minutes of receiving 2-3 votes. I have no affiliation with YC, nor are my submissions.
chickenzzzzu•2h ago
Sounds more like a cattle auction than a place where partnerships are formed to deliver meaningful value to consumers.
echelon•2h ago
This is market efficiency. That's the very definition of delivering meaningful value.

The more efficient the capital allocation is, the greater the benefit to the whole market.

The only weirdness is that these are companies led by founders with "personalities". That's a weird variable to correct for.

chickenzzzzu•1h ago
How much more efficient is it to allocate capital to a group individuals who have been educated at institutions that produce an increasingly rigid type of thinker, who are all making the same hype driven products?
orsorna•2h ago
If you aren't already connected beforehand these rituals are a waste of time.
chickenzzzzu•1h ago
Painfully true!
soared•2h ago
https://billchambers.me/about/

This page on mobile made me laugh.

But otherwise, makes sense - get rid off all the flash and this is what demo days/etc are about

ecb_penguin•1h ago
I didn't realize how poorly YC startups were managed until I joined one.

It's mostly a ponzi scheme, where unprofitable companies pass their books off to non-technical investors that love hype.

> 1 slide. 1 minute pitches. 1 speaker.

This is why YC has such poor results. They're more interested in "1 slide" vs actual financials.

> Vertical AI was all over the place. Cursor for X was also prevalent.

No thought leadership at all.

HPMOR•1h ago
YC is very similar to a university. There is a small group of companies that create the reputation on which everybody lives. The results are extremely exponentially distributed. So yes, it makes sense you joining some arbitrary YC startup was a bad experience. The average performance of a YC company is __phenomenal__, the median performance is poor.
ecb_penguin•1h ago
> YC is very similar to a university

No it's not. It's not even remotely comparable.

> There is a small group of companies that create the reputation on which everybody lives

Also called luck. YC investors cannot articulate what made them successful, or else they'd have better results.

Universities have decades of sustained output. They are not comparable.

> you joining some arbitrary YC startup was a bad experience

I am at a YC unicorn. We are one of YC's most successful startups.

> The average performance of a YC company is __phenomenal__

It absolutely is not. It doesn't even beat the SP500. The __vast__ majority of YC companies are unprofitable failures.

You could flip a coin and beat YC.

HPMOR•1h ago
You're wrong. YC's returns far exceed the returns of the S&P 500. This is why LP's throw money at YC's funds. And secondly, I've been through YC. It is very apparent after many many group office hours, which startups are likely to be in the 6% unicorn group and which are not. You are correct the vast majority of YC companies are failures. However this does not preclude that the *mean* of YC outcomes is very very good. If Bill Gates were in a room with me and my friends the MEAN net worth would be tens of billions of dollars, the median would be less than a $100k.
ecb_penguin•58m ago
> You're wrong

You thought YC was comparable to a university

> This is why LP's throw money at YC's funds

Yeah, we know hedge funds that throw money at things usually beat the SP500.

> It is very apparent after many many group office hours, which startups are likely to be in the 6% unicorn group and which are not

Group office hours and not actual revenue. Proving my point!

> However this does not preclude that the mean of YC outcomes is very very good

It absolutely does, lmao.

> If Bill Gates were in a room with me and my friends the MEAN net worth would be tens of billions of dollars, the median would be less than a $100k.

Yes, one person being successful, the rest failures would skew the results! Way to prove the point I'm making, lmao.

jasonfrost•27m ago
You agreed with his rephrasing of his original assertion
pyb•1h ago
"2 on 20" Am I understanding correctly? If every YC startup is shooting for a 20 million USD valuation, that's amazing. Even if not all will make it.
HPMOR•1h ago
Nobody ever pays too much or too little for a startup. Because either you __really__ make it, or you don't. It is statistically more likely for your YC startup to be worth more than $1 billion, than it is to be acquired in a cash deal for $80mm. It is because of this fact that $2/22mm is not unreasonable. However, most of the hot companies during the batch will fetch $3-4mm/30-40mm valuation. The average raise during demo day is close to ~$3mm, the median <$800k.

Uv and Ray: Pain-Free Python Dependencies in Clusters

https://www.anyscale.com/blog/uv-ray-pain-free-python-dependencies-in-clusters
1•robertnishihara•22s ago•0 comments

Timeline of US Class I Railroads Since 1977

https://en.wikipedia.org/wiki/Timeline_of_Class_I_railroads_(1977%E2%80%93present)
1•brudgers•7m ago•0 comments

We should train AI in space [pdf]

https://starcloudinc.github.io/wp.pdf
1•andsoitis•7m ago•0 comments

British man charged by US with leading hacking scheme causing millions in damage

https://news.sky.com/story/british-man-charged-by-us-with-leading-hacking-scheme-and-causing-millions-in-damages-13389185
1•austinallegro•8m ago•0 comments

Apple Just Patented an Image Sensor with 20 Stops of Dynamic Range

https://ymcinema.com/2025/06/25/apple-just-patented-an-image-sensor-with-20-stops-of-dynamic-range/
1•consumer451•9m ago•1 comments

Type Theory and Themes in Philosophical Logic – Greg Restall – TYPES 2025

https://consequently.org/presentation/2025/type-theory-and-phil-logic-types/
1•matt_d•11m ago•0 comments

Show HN: Sick of emailing yourself stuff? me too

https://github.com/sirbread/sink
2•sirbread•12m ago•0 comments

AI's Cognitive Implications: The Decline of Our Thinking Skills?

https://www.ie.edu/center-for-health-and-well-being/blog/ais-cognitive-implications-the-decline-of-our-thinking-skills/
1•jruohonen•14m ago•0 comments

Elon Musk's A.I. Company Faces Lawsuit over Gas-Burning Turbines

https://www.nytimes.com/2025/06/17/climate/naacp-musk-xai-supercomputer-lawsuit.html
1•1vuio0pswjnm7•17m ago•0 comments

How much energy does your AI prompt useuse

https://www.wsj.com/tech/ai/ai-prompt-video-energy-electricity-use-046766d6
3•1vuio0pswjnm7•24m ago•0 comments

Rick Perry's Fermi to Build Giant AI Nuclear, NatGas, Solar Center

https://www.hartenergy.com/exclusives/rick-perrys-fermi-build-giant-ai-nuclear-natgas-solar-center-213392
1•mpweiher•25m ago•0 comments

Protecting Science: TIB Builds Dark Archive for ArXiv

https://blog.tib.eu/2025/05/14/protecting-science-tib-builds-dark-archive-for-arxiv/
1•lentoutcry•25m ago•0 comments

Modelling API rate limits as diophantine inequalities

https://vivekn.dev/blog/rate-limit-diophantine
2•viveknathani_•26m ago•1 comments

Airtable Has Pivoted to Vibe Coding

https://www.airtable.com/
3•sh_tomer•26m ago•0 comments

Kaito – Kubernetes AI toolchain operator

https://github.com/kaito-project/kaito
1•eh8•27m ago•0 comments

Big accounting firms fail to track AI impact on audit quality, says regulator

https://www.ft.com/content/ee5be0d2-6d7b-432b-b778-1f7e471d8145
2•bishopsmother•27m ago•0 comments

AI Voice Agent Blog

https://www.vidnova.ai/
1•plussmart2018•28m ago•0 comments

A case for courage, when speaking of AI danger

https://www.lesswrong.com/posts/CYTwRZtrhHuYf7QYu/a-case-for-courage-when-speaking-of-ai-danger
1•ibobev•29m ago•0 comments

Scientists invent photosynthetic material that sucks CO2 out of the atmosphere

https://www.livescience.com/technology/engineering/scientists-invent-photosynthetic-living-material-that-sucks-co2-out-of-the-atmosphere
2•bdev12345•30m ago•0 comments

The Secret Rules of the Terminal

https://jvns.ca/blog/2025/06/24/new-zine--the-secret-rules-of-the-terminal/
2•ibobev•31m ago•0 comments

Sprout: CLI tool to automate Git worktree and Docker Compose workflows

https://github.com/SecDev-Lab/sprout
1•handfuloflight•32m ago•0 comments

Bitsets match regular expressions, compactly (2013)

https://pvk.ca/Blog/2013/06/23/bitsets-match-regular-expressions/
2•Bogdanp•33m ago•0 comments

Windows 11 PCs performance details

https://learn.microsoft.com/en-us/legal/windows/windows11-performance-claims
1•XzetaU8•37m ago•0 comments

Show HN: Free decision calculator to avoid spreadsheet hell

https://www.kleardecision.com/
2•kathysledge•37m ago•0 comments

Show HN: Free tool to resize ChatGPT images to perfect 16:9

https://www.aithumbnail.so/tools/chatgpt-image-resizer
3•sachou•38m ago•1 comments

Show HN: Checkpoint K8s pods transparently (plain CPU or GPU accelerated) [video]

https://www.youtube.com/watch?v=K9yY6_2255Y
1•qnib•40m ago•0 comments

ASML launches science project in China to discover new talent in lithography

https://www.scmp.com/tech/tech-war/article/3315938/asml-launches-science-project-china-discover-new-talent-lithography
1•Ozarkian•41m ago•0 comments

Bettina Heim

https://en.wikipedia.org/wiki/Bettina_Heim
1•jayhoon•42m ago•0 comments

Parameterized types in C using the new tag compatibility rule

https://nullprogram.com/blog/2025/06/26/
2•ingve•42m ago•0 comments

China offers an alternative to Western 'technofeudalism'

https://www.scmp.com/opinion/china-opinion/article/3314964/china-offers-alternative-western-technofeudalism
3•thunderbong•44m ago•1 comments