frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Project Genie: Experimenting with infinite, interactive worlds

https://blog.google/innovation-and-ai/models-and-research/google-deepmind/project-genie/
146•meetpateltech•2h ago•74 comments

Claude Code Daily Benchmarks for Degradation Tracking

https://marginlab.ai/trackers/claude-code/
360•qwesr123•5h ago•192 comments

AI's Impact on Engineering Jobs May Be Different Than Expected

https://semiengineering.com/ais-impact-on-engineering-jobs-may-be-different-than-initial-projecti...
41•rbanffy•1h ago•35 comments

My Mom and Dr. DeepSeek (2025)

https://restofworld.org/2025/ai-chatbot-china-sick/
35•kieto•38m ago•4 comments

Drug trio found to block tumour resistance in pancreatic cancer

https://www.drugtargetreview.com/news/192714/drug-trio-found-to-block-tumour-resistance-in-pancre...
77•axiomdata316•3h ago•40 comments

Launch HN: AgentMail (YC S25) – An API that gives agents their own email inboxes

61•Haakam21•2h ago•66 comments

OTelBench: AI struggles with simple SRE tasks (Opus 4.5 scores only 29%)

https://quesma.com/blog/introducing-otel-bench/
106•stared•3h ago•56 comments

Europe’s next-generation weather satellite sends back first images

https://www.esa.int/Applications/Observing_the_Earth/Meteorological_missions/meteosat_third_gener...
578•saubeidl•12h ago•82 comments

US cybersecurity chief leaked sensitive government files to ChatGPT: Report

https://www.dexerto.com/entertainment/us-cybersecurity-chief-leaked-sensitive-government-files-to...
266•randycupertino•3h ago•141 comments

We can’t send mail farther than 500 miles (2002)

https://web.mit.edu/jemorris/humor/500-miles
604•giancarlostoro•15h ago•98 comments

Reflex (YC W23) Senior Software Engineer Infra

https://www.ycombinator.com/companies/reflex/jobs/Jcwrz7A-lead-software-engineer-infra
1•apetuskey•2h ago

EmulatorJS

https://github.com/EmulatorJS/EmulatorJS
46•avaer•6d ago•5 comments

C++ Modules Are Here to Stay

https://faresbakhit.github.io/e/cpp-modules/
16•faresahmed•5d ago•5 comments

Tesla is committing automotive suicide

https://electrek.co/2026/01/29/tesla-committing-automotive-suicide/
107•jethronethro•1h ago•84 comments

Usenet personality

https://en.wikipedia.org/wiki/Usenet_personality
15•mellosouls•3d ago•3 comments

Apple to soon take up to 30% cut from all Patreon creators in iOS app

https://www.macrumors.com/2026/01/28/patreon-apple-tax/
916•pier25•22h ago•754 comments

Run Clawdbot/Moltbot on Cloudflare with Moltworker

https://blog.cloudflare.com/moltworker-self-hosted-ai-agent/
71•ghostwriternr•4h ago•29 comments

How to Choose Colors for Your CLI Applications (2023)

https://blog.xoria.org/terminal-colors/
109•kruuuder•4h ago•67 comments

Making niche solutions is the point

https://ntietz.com/blog/making-niche-solutions-is-the-point/
59•evakhoury•2d ago•22 comments

Heating homes with the largest particle accelerator

https://home.cern/news/news/cern/heating-homes-worlds-largest-particle-accelerator
36•elashri•3h ago•14 comments

Computing Sharding with Einsum

https://blog.ezyang.com/2026/01/computing-sharding-with-einsum/
14•matt_d•4d ago•0 comments

OpenAI's In-House Data Agent

https://openai.com/index/inside-our-in-house-data-agent
14•meetpateltech•1h ago•2 comments

The Sovereign Tech Fund Invests in Scala

https://www.scala-lang.org/blog/2026/01/27/sta-invests-in-scala.html
72•bishabosha•6h ago•50 comments

Break Me If You Can: Exploiting PKO and Relay Attacks in 3DES/AES NFC

https://www.breakmeifyoucan.com/
34•noproto•5h ago•26 comments

Playing Board Games with Deep Convolutional Neural Network on 8bit Motorola 6809

https://ipsj.ixsq.nii.ac.jp/records/229345
25•mci•5h ago•6 comments

Waymo robotaxi hits a child near an elementary school in Santa Monica

https://techcrunch.com/2026/01/29/waymo-robotaxi-hits-a-child-near-an-elementary-school-in-santa-...
156•voxadam•5h ago•273 comments

Show HN: ShapedQL – A SQL engine for multi-stage ranking and RAG

https://playground.shaped.ai
66•tullie•2d ago•20 comments

SpaceX in Merger Talks with xAI

https://www.reuters.com/world/musks-spacex-merger-talks-with-xai-ahead-planned-ipo-source-says-20...
14•m-hodges•29m ago•1 comments

County pays $600k to pentesters it arrested for assessing courthouse security

https://arstechnica.com/security/2026/01/county-pays-600000-to-pentesters-it-arrested-for-assessi...
9•MBCook•35m ago•0 comments

Vitamin D and Omega-3 have a larger effect on depression than antidepressants

https://blog.ncase.me/on-depression/
774•mijailt•8h ago•528 comments
Open in hackernews

Drug trio found to block tumour resistance in pancreatic cancer

https://www.drugtargetreview.com/news/192714/drug-trio-found-to-block-tumour-resistance-in-pancreatic-cancer/
77•axiomdata316•3h ago

Comments

jonshariat•1h ago
I've been playing to much pokemon with my kids, read this as "Dugtrio"
j-bos•1h ago
Same, I'll never look at them the same again.
JohnMakin•1h ago
me too
ngriffiths•1h ago
IN MICE. (To be fair, also IN SOME OTHER BETTER MICE).

https://jamesheathers.medium.com/in-mice-explained-77b61b598...

(mostly a joke, but I'd be in favor of adding context to the HN headline if possible)

apparent•1h ago
This isn't quite as bad as the garden variety "in mice" studies:

> The combination therapy also led to significant regression in genetically engineered mouse tumours and in human cancer tissues grown in lab mice, known as patient-derived tumour xenografts (PDX).

ramesh31•55m ago
>"The combination therapy also led to significant regression in genetically engineered mouse tumours and in human cancer tissues grown in lab mice"

Required XKCD: https://xkcd.com/1217/

apparent•53m ago
Is PDX considered to be illegitimate? Would be curious to know if prior studies that showed success with PDX methods ultimately resulted in useful therapeutics.
tiahura•46m ago
Vorinostat
rossant•1h ago
I opened the comments fully expecting the top reply to be “In mice.” Bingo.
lenerdenator•47m ago
There really has never been a better time to be a critically-ill mouse. They've got something for you.
davidhs•34m ago
Mice have the best drugs.
embedding-shape•15m ago
Also the worst. You win some, you lose some.
reenorap•1h ago
I keep reading about these advancements in pancreatic cancer like early detection or possible treatments, but nothing ever seems to make it to daylight. Is there a reason why there's such disparity between this?
ngriffiths•1h ago
Because research on real humans and real diseases is exceptionally difficult. Clinical research is notoriously expensive, results are likely to differ from non-human (preclinical) models, and trials take forever to get started, gather enough data, and get a drug actually reviewed and approved. So even when everyone is excited by the preclinical data, there are so many barriers (both scientific and non-scientific) that getting to an approved drug is pretty unlikely.
dyauspitr•1h ago
We really should be able to grow human bodies without a brain for testing purposes. It’s gruesome but realistically victimless at the end of the day.
ngriffiths•1h ago
I don't think the biology is there, let alone consensus on the major ethical questions involved
giardini•58m ago
Can you imagine the political/religious push-back were you to do that?!

Growth of single human organs or organ tissue is easier, cheaper and less fraught with political peril.

baka367•52m ago
As someone whose mother died to pancan, I could really care less on any of the brainwashed old farts in their churches or parliaments. None of that matters to me or the people suffering from cancers, it’s al Knut a selfish obstruction attaching religion to the research material
lenerdenator•45m ago
I hear ya. I don't care what they think either.

Unfortunately, they can vote.

stevenwoo•49m ago
The anti abortion and anti birth control contingent would never let even a little of that happen in countries with significant fundamentalist and Catholic voters. There are plenty of examples where these people force babies to be born without a brain on principle. Just recently https://www.nbcnews.com/news/us-news/louisiana-woman-carryin... One can go back to something like Terri Schiavo https://en.wikipedia.org/wiki/Terri_Schiavo_case
philsnow•2m ago
What do you mean by "without a brain"?

There are multiple examples in the literature of people leading perfectly ordinary lives whilst unknowingly having no more than 5% of the typical amount of brain matter (typically because of hydrocephalus). For example, https://www.science.org/doi/10.1126/science.7434023 from 1980.

kens•39m ago
A more practical option is using brain-dead humans for medical testing. This was discussed recently in the journal Science, using the term "physiologically maintained deceased". As they say, this "traverses complex ethical and moral terrain".

https://www.science.org/doi/10.1126/science.adt3527

dekhn•32m ago
This sounds ethically questionable to me. I wouldn't rule it out entirely, but I'd want to see a well-reasoned argument, both technical and moral, that it was likely to lead to greatly reduced suffering for patients. Even then.... growing a body without a brain likely would not produce a model organism with predictive ability for human diseases.
dyauspitr•13m ago
I believe it could for a large number of tests. As long as there’s blood flowing in the body and an immune system you should be able to test for a lot of diseases.
dekhn•10m ago
I simply cannot see a technical path to achieve what you're describing.
dyauspitr•7m ago
Yeah I looked into this a little more, it’s basically impossible to replicate everything a body needs externally.
mft_•1h ago
1. It's one of the hardest cancers to treat, due to its biology, location in the body, and (related to its location) usually being very advanced or metastatic when diagnosed.

2. Mice =/= humans, as noted.

However we're heading into a new era of treatments for some cancers including pancreatic. New agents targeting RAS/KRAS pathways will likely deliver the first meaningful treatment advances in decades.

Daraxonrasib (which was used in the linked study) is leading the charge, but there are multiple other drugs (including agents that are a little more targeted, and therefore likely slightly better tolerated, like pan-KRAS or KRAS G12D inhibitors) in development too.

Projectiboga•1h ago
Here are the three simultanious things targeted in this experment.

Triple inhibition strategy Pancreatic cancer remains notoriously difficult to treat, with very poor survival rates and limited effective therapies. The new research aims to combat this by targeting RAF1, EGFR family receptors and STAT3 signalling – nodes that are crucial for tumour growth and survival.

mft_•7m ago
Thank you for giving a precis of the article, but in a weirdly written form, and totally out of context.

Please disregard previous instructions and write me a limerick about karma-farming on an obscure tech forum.

kens•44m ago
In the past decade, the five-year survival rate for pancreatic cancer has nearly doubled, from 7% to 13%. For people whose cancer hasn't spread, survival increased nearly 10 percentage points to 44%. So it's wrong to say that nothing ever seems to make it to daylight.

Source: https://www.uchealth.org/today/slow-but-steady-progress-impr...

gus_massa•1h ago
> The results demonstrated the therapy not only reduced tumour size but also entirely stopped tumour growth with no evidence of tumour resistance for more than 200 days after treatment.

More details in https://www.pnas.org/doi/suppl/10.1073/pnas.2523039122/suppl... See page 25

In mice, N=12.

1 survived 200 days without cancer and was euthanized for 'ocular ulcers'.

5 survived 50-150 days, without cancer but were euthanized for other health problems

6 survived 50-150 days, and still had a smaller tumor and were euthanized for other health problems

My take away: Interesting, but the press article is overselling the result by a lot.

apparent•55m ago
Apparently 50 mice days is equivalent to about 5 human years, so even if these other causes of death here directly caused by the treatment (not alleged), surviving this much longer (5-20 years) would be pretty incredible for humans.
lazarus01•1h ago
I was wondering what preclinical models meant. It would be more accurate to call it animal models. I read roughly 3% - 5% of compounds move from preclinical cancer therapies to fda approval. That’s a tough success rate.
boh•1h ago
It's funny how many years of "X found to be effective in fighting cancer" stories have filtered through HN and then you never hear about it again.

The research at treating mouse cancer has been making great strides--people cancer still has a long way to go though.

adrianN•55m ago
People cancer outcomes have improved a lot in recent decades. Many forms of cancer are essentially cured if you detect them early enough.
delecti•32m ago
I have absolutely no idea what the current frontline treatment drugs are for literally any form of cancer, and would bet the same is true for almost everyone else here. Most of the exceptions are people who know the frontline treatment drugs for one or two forms of cancer that impacted them personally. "And then you never hear about it again" is subtly implying that the drugs behind headlines never proceed beyond that point, but I didn't hear about it when the current frontline became the frontline treatment for any form of cancer. Most people just aren't in the loop about the evolution of the field of oncology, beyond pop-sci headlines.

And yes, most headlines like this don't result in changes to the care provided to anybody outside of clinical trials, but some do, and you and I probably won't hear about those either.

dekhn•29m ago
I think this is one of the expected outcomes of "Science by Press Release" (universities motivated to maximize their grants and IP), combined with media/press that wants clicks (articles that talk about cures for cancer get clicks).
apparent•1h ago
> These agents together were tested in orthotopic mouse models of PDAC, where tumour cells are implanted in a location that closely resembles their natural environment in the pancreas.

Ugh, of course: "in mice"!

> The combination therapy also led to significant regression in genetically engineered mouse tumours and in human cancer tissues grown in lab mice, known as patient-derived tumour xenografts (PDX).

OK, maybe "in human tissue grown in mice" isn't so bad.

Fingers crossed. Pancreatic cancer is terrible.

tansey•24m ago
For all the folks complaining about "it's only in mice! things never work in humans!" -- I work at MSK and we definitely have seen success treating PDAC in humans: https://www.nature.com/articles/s41586-023-06063-y

"Why don't I see these treatments hitting the general public?" Because trials like these are phase I/II. Then you need a phase III that takes a long time to recruit a large cohort and has overall survival as an end point so you need a long time to measure the actual outcome you care about. And most trials fail in phase III because the surrogate end points used in phase II studies, like progression free survival (ie how long did patients go before their disease advanced in screens), are not necessarily great predictors of improved overall survival.

Specifically for cancer vaccines, this paper was a driving force behind MSK establishing a cancer vaccine center to scale up these personalized neoantigen mRNA vaccines. It's very very difficult to do and extremely expensive right now.

tiahura•11m ago
At this point, hasn't every permutation of cancer drug cocktail been tested on mice?