frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

Open in hackernews

How to Do Ambitious Research in the Modern Era [video]

https://www.youtube.com/watch?v=w7DVlI_Ztq8
52•surprisetalk•1d ago

Comments

gsf_emergency•1d ago
For those like me who weren't familiar with the speaker's research, this could be interesting:

https://hn.algolia.com/?dateRange=all&page=2&prefix=true&que...

notarobot123•22h ago
I love the idea of "industrial research" that's somewhere between the castle-in-the-sky thinking of academia and cold rationalism of market-oriented product development.

The fact that Ink & Switch researchers are focusing on "authentic use" without the pressure to find product-market-fit means they can effectively scout the terrain and signpost the fruitful paths.

Imagine if today's tech monopolies were compelled to invest in this kind of research (just like AT&T was). It would create a hot-bed of innovation and new avenues would open up for startups to explore. I wish we had more research groups like this.

bitmasher9•21h ago
Everyone tries to replicate bell labs, but no one has been close. Simply spending the money isn’t enough to get there.
notarobot123•20h ago
This is exactly why the "hollywood model" of fast iterative cycles of focused peer-evaluated research is such an exciting development.

I think Ink & Switch are presenting a model for industrial research that could absolutely be replicated by other teams following a similar style of output-oriented work.

kjkjadksj•17h ago
I feel like that would lead to burnout of the staff working at that pace for a career.

What would be interesting if we could be empowered to go back to that say 19th century idea of the reasearcher. You know, someone with all their wants and needs seemingly acceptably met, tinkering on some obsession they have in their workshop/lab on their own property or out collecting data around the world. Only with modern technology and ideas of course.

bitmasher9•9h ago
I think it’s fair that a 19th century style of life doesn’t exist in the 21st century.
jonstewart•21h ago
I would love to know who the speaker is without having to watch a video.
DavidPiper•21h ago
I agree with you, but also 0:37.

> Peter has worked as an Arctic oceanographer, a game developer, he's wrangled millions of Postgres databases and he's now executive director of Ink & Switch where he plots the future of tools for thought.

DavidPiper•20h ago
Interesting that he's essentially saying all the places where this kind of deep skunkwork / research is "supposed" to happen (startups, universities, government, etc) have all prioritised or been captured by other incentives.

Weirdly almost feels analogous to the whole "loss of a third space" conversation: we're losing places that can actually (afford to) invest in deep research and exploration.

gonzo41•20h ago
I currently work for a research oriented organization. The incentive structures for the research staff, firstly atomize the staff and secondly directly work against the organization goals of being able to get impactful things done.

Beware of Fast-Math

https://simonbyrne.github.io/notes/fastmath/
81•blobcode•3h ago•23 comments

Photos taken inside musical instruments

https://www.dpreview.com/photography/5400934096/probe-lenses-and-focus-stacking-the-secrets-to-incredible-photos-taken-inside-instruments
635•worik•13h ago•29 comments

Gradients Are the New Intervals

https://www.mattkeeter.com/blog/2025-05-14-gradients/
29•surprisetalk•3h ago•0 comments

Valkey Turns One: Community fork of Redis

https://www.gomomento.com/blog/valkey-turns-one-how-the-community-fork-left-redis-in-the-dust/
185•cebert•11h ago•69 comments

Cerebras achieves 2,500T/s on Llama 4 Maverick (400B)

https://www.cerebras.ai/press-release/maverick
48•ByteAtATime•6h ago•27 comments

Simpler Backoff

https://commaok.xyz/post/simple-backoff/
51•todsacerdoti•5h ago•13 comments

Surprisingly fast AI-generated kernels we didn't mean to publish yet

https://crfm.stanford.edu/2025/05/28/fast-kernels.html
275•mfiguiere•14h ago•77 comments

The ‘white-collar bloodbath’ is all part of the AI hype machine

https://www.cnn.com/2025/05/30/business/anthropic-amodei-ai-jobs-nightcap
432•lwo32k•20h ago•727 comments

Webb telescope helps refines Hubble constant, suggesting resolution rate debate

https://phys.org/news/2025-05-webb-telescope-refines-hubble-constant.html
3•pseudolus•3d ago•0 comments

Beating Google's kernelCTF PoW using AVX512

https://anemato.de/blog/kctf-vdf
286•anematode•17h ago•88 comments

What's working for YC companies since the AI boom

https://jamesin.substack.com/p/whats-working-for-yc-companies-since
73•jseidel•19h ago•43 comments

AccessOwl (YC S22) is hiring an AI TypeScript Engineer to connect 100s of SaaS

https://www.ycombinator.com/companies/accessowl/jobs/hfWAhVp-ai-enabled-senior-software-engineer-typescript-focus
1•mathiasn•3h ago

Reverse engineering of Linear's sync engine

https://github.com/wzhudev/reverse-linear-sync-engine
102•flashblaze•2d ago•12 comments

Randomness Requirements for Security

https://datatracker.ietf.org/doc/html/rfc4086
12•mooreds•2d ago•0 comments

Show HN: Icepi Zero – The FPGA Raspberry Pi Zero Equivalent

https://github.com/cheyao/icepi-zero
161•Cyao•2d ago•39 comments

Revenge of the Chickenized Reverse-Centaurs

https://pluralistic.net/2022/04/17/revenge-of-the-chickenized-reverse-centaurs/
107•GreenWatermelon•2d ago•32 comments

Microsandbox: Virtual Machines that feel and perform like containers

https://github.com/microsandbox/microsandbox
301•makeboss•20h ago•145 comments

Mary Meeker's first Trends report since 2019, focused on AI

https://www.bondcap.com/reports/tai
141•kjhughes•14h ago•60 comments

C++ to Rust Phrasebook

https://cel.cs.brown.edu/crp/
65•wcrichton•12h ago•5 comments

Show HN: MCP Defender – OSS AI Firewall for Protecting MCP in Cursor/Claude etc

https://mcpdefender.com
36•gsundeep•1d ago•32 comments

AI Responses May Include Mistakes

https://www.os2museum.com/wp/ai-responses-may-include-mistakes/
153•userbinator•4h ago•116 comments

Systems Correctness Practices at Amazon Web Services

https://cacm.acm.org/practice/systems-correctness-practices-at-amazon-web-services/
331•tanelpoder•21h ago•120 comments

Java Virtual Threads Ate My Memory: A Web Crawler's Tale of Speed vs. Memory

https://dariobalinzo.medium.com/virtual-threads-ate-my-memory-a-web-crawlers-tale-of-speed-vs-memory-a92fc75085f6
72•dariobalinzo•1d ago•24 comments

Cap: Lightweight, modern open-source CAPTCHA alternative using proof-of-work

https://capjs.js.org/
135•tiagorangel•17h ago•92 comments

How large should your sample size be?

https://vickiboykis.com/2015/08/04/how-large-should-your-sample-size-be/
27•sebg•3d ago•19 comments

Ray Tracing in J

https://idle.nprescott.com/2020/ray-tracing-in-j.html
65•todsacerdoti•2d ago•9 comments

The Darwin Gödel Machine: AI that improves itself by rewriting its own code

https://sakana.ai/dgm/
182•birriel•22h ago•179 comments

Every 5x5 Nonogram

https://pixelogic.app/every-5x5-nonogram
60•eieio•10h ago•24 comments

Doge cuts to USAid blamed for 300k deaths – most of them children

https://www.thetimes.com/us/american-politics/article/usaid-doge-deaths-children-cuts-7nb83dfkp
15•mnewme•2h ago•4 comments

Silicon Valley finally has a big electronics retailer again: Micro Center opens

https://www.microcenter.com/site/mc-news/article/micro-center-santa-clara-photos.aspx
234•modeless•11h ago•121 comments