<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: kevinalexbrown</title><link>https://news.ycombinator.com/user?id=kevinalexbrown</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 21 Nov 2026 00:29:50 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=kevinalexbrown" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by kevinalexbrown in "The Dark Night of Mathematics"]]></title><description><![CDATA[
<p>It’s ok to mourn the loss of a frontier and also support the thing driving that loss. The way it looks, the generation of mathematical results will become less and less of a human endeavor. There’s a way that’s sad even if on the whole it is a huge leap for society. This was a great article, and I applaud the author for acknowledging this head on.</p>
]]></description><pubDate>Sat, 25 Jul 2026 17:52:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=49049877</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=49049877</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49049877</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Claude Fable 5"]]></title><description><![CDATA[
<p>"tell me about biology" -> "Switched to Opus 4.8"</p>
]]></description><pubDate>Tue, 09 Jun 2026 19:55:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=48466766</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=48466766</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48466766</guid></item><item><title><![CDATA[The Patient Is Not a Document: Moving from LLMs to a World Model for Oncology]]></title><description><![CDATA[
<p>Article URL: <a href="https://blog.standardmodel.bio/p/the-patient-is-not-a-document-moving-f88">https://blog.standardmodel.bio/p/the-patient-is-not-a-document-moving-f88</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46508139">https://news.ycombinator.com/item?id=46508139</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 06 Jan 2026 02:44:43 +0000</pubDate><link>https://blog.standardmodel.bio/p/the-patient-is-not-a-document-moving-f88</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=46508139</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46508139</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Ask HN: What Are You Working On? (December 2025)"]]></title><description><![CDATA[
<p>Building the fndamental human patient representation: <a href="https://standardmodelbio.substack.com/p/the-patient-is-not-a-document-moving" rel="nofollow">https://standardmodelbio.substack.com/p/the-patient-is-not-a...</a><p>The patient is not a document - multimodal foundation models for biomedicine. JEPA's working well.</p>
]]></description><pubDate>Mon, 15 Dec 2025 00:46:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=46268921</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=46268921</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46268921</guid></item><item><title><![CDATA[The Patient Is Not a Document: Moving from LLMs to a World Model for Oncology]]></title><description><![CDATA[
<p>Article URL: <a href="https://blog.standardmodel.bio/p/the-patient-is-not-a-document-moving">https://blog.standardmodel.bio/p/the-patient-is-not-a-document-moving</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46165097">https://news.ycombinator.com/item?id=46165097</a></p>
<p>Points: 7</p>
<p># Comments: 1</p>
]]></description><pubDate>Fri, 05 Dec 2025 18:20:02 +0000</pubDate><link>https://blog.standardmodel.bio/p/the-patient-is-not-a-document-moving</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=46165097</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46165097</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Ask HN: Who is hiring? (November 2025)"]]></title><description><![CDATA[
<p>Standard Model Bio | various | Full-Time | ONSITE SF or Philadelphia | Hybrid<p>we're building the standard model for bio. We're doing for biology what mathematics did for physics.<p>papers this month: 
genomics - <a href="https://arxiv.org/abs/2509.25573" rel="nofollow">https://arxiv.org/abs/2509.25573</a>
protein-language: <a href="https://arxiv.org/abs/2509.22853" rel="nofollow">https://arxiv.org/abs/2509.22853</a>
longitudinal ehr: <a href="https://arxiv.org/abs/2509.25591" rel="nofollow">https://arxiv.org/abs/2509.25591</a><p>hello world: <a href="https://standardmodelbio.substack.com/p/introducing-standard-model-biomedicine" rel="nofollow">https://standardmodelbio.substack.com/p/introducing-standard...</a><p>we're humble and ambitious - we want to be the quiet backbone of biomedical ai but we want none of the glory of the final applications.<p>kevin@standardmodel.bio</p>
]]></description><pubDate>Mon, 03 Nov 2025 20:42:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=45804169</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=45804169</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45804169</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Ask HN: What are you working on? (October 2025)"]]></title><description><![CDATA[
<p>building the standard model for bio: <a href="https://standardmodelbio.substack.com/p/introducing-standard-model-biomedicine" rel="nofollow">https://standardmodelbio.substack.com/p/introducing-standard...</a><p>We're pretty jazzed.</p>
]]></description><pubDate>Sun, 12 Oct 2025 22:43:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=45562726</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=45562726</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45562726</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Japanese scientists develop vaccine to eliminate cells behind aging"]]></title><description><![CDATA[
<p>Yes and this response has also become a trope. It’s generically universal to nearly any early result.</p>
]]></description><pubDate>Sun, 12 Dec 2021 16:48:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=29531101</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=29531101</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29531101</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "“AI promised to revolutionize radiology but so far its failing”"]]></title><description><![CDATA[
<p>The QZ article is narrowly correct but widely misleading. It almost willfully ignores the momentum and direction.<p>In reality, radiologists will not be summarily replaced one day. They will get more and more productive as tools extend their reach. This can occur even as the number of radiologists increases.<p>Here's a recent example where Hinton was right in concept: recent AI work for lung cancer detection made radiologists perform better in an FDA 510k clearance.<p><i>20 readers reviewed all of 232 cases using both a second-reader as well as a concurrent first reader workflows. Following the read according to both workflows, five expert radiologists reviewed all consolidated marks. The reference standard was based on reader majority (three out of five) followed by expert adjudication, as needed. As a result of the study’s truthing process, 143 cases were identified as including at least one true nodule and 89 with no true nodules. All endpoints of the analyses were satisfactorily met. These analyses demonstrated that all readers showed a significant improvement for the detection of pulmonary nodules (solid, part-solid and ground glass) with both reading workflows. </i><p><a href="https://www.accessdata.fda.gov/cdrh_docs/pdf20/K203258.pdf" rel="nofollow">https://www.accessdata.fda.gov/cdrh_docs/pdf20/K203258.pdf</a><p>(I am proud to have worked with others on versions of the above, but do not speak for them or the approval, etc)<p>The AI revolution in medicine is here. That is not in dispute by most clinicians in training now, nor, from all signs, by the FDA. Not everyone is making use of it yet, and not all of it is perfect (as with radiologists - just try to get a clean training set). But the idea that machine learning/ai is overpromising is like criticizing Steve Jobs in 2008 for overpromising the iphone by saying it hasn't totally changed your life yet. Ok.</p>
]]></description><pubDate>Mon, 07 Jun 2021 16:15:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=27424287</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=27424287</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=27424287</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Ask HN: Who is hiring? (February 2020)"]]></title><description><![CDATA[
<p>Siemens Healthineers Imaging Intelligence | <a href="https://www.siemens-healthineers.com" rel="nofollow">https://www.siemens-healthineers.com</a> | Malvern, PA (Greater Philadelphia) | INTERNS | Onsite Our R&D group delivers medical image/text tools (e.g. deep learning, NLP, etc) for medical data analysis. We are well recognized for delivering cutting-edge intelligent solutions to Siemens 3D workstations and medical imaging scanners. Our group also has strong publication record in top tier journals and conferences, and several Siemens "inventor of the year" award recipients.
We offer well-paid internships lasting >= 3 months, with independent moonshot projects.<p>Responsibilities: · Contribute to research projects to develop intelligent solutions for medical imaging and text analytics · Conduct fast prototyping, feasibility studies for exploratory clinical research · Support the productization of research prototypes<p>We look for: · Strong research capability in computer vision, machine learning, text analytics and medical image analysis, proven by publications in journals/conferences. · Research experience in image/text analytics using large scale, weakly supervised / unsupervised learning algorithms · Research experience in medical image/text analysis of different modalities (CT, MRI, PET, medical reports etc.)<p>Email: Kevin.Brown@siemens-healthineers.com</p>
]]></description><pubDate>Tue, 04 Feb 2020 14:25:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=22235820</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=22235820</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=22235820</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Ask HN: Who is hiring? (January 2020)"]]></title><description><![CDATA[
<p>Siemens Healthineers Imaging Intelligence | <a href="https://www.siemens-healthineers.com" rel="nofollow">https://www.siemens-healthineers.com</a> | Malvern, PA (Greater Philadelphia) | INTERNS | Onsite
Our R&D group delivers medical image/text tools (e.g. deep learning, NLP, etc) for medical data analysis. We are well recognized for delivering cutting-edge intelligent solutions to Siemens 3D workstations and medical imaging scanners. Our group also has strong publication record in top tier journals and conferences, and several Siemens "inventor of the year" award recipients.<p>We offer well-paid internships lasting >= 3 months, with independent moonshot projects.<p>Responsibilities: · Contribute to research projects to develop intelligent solutions for medical imaging and text analytics · Conduct fast prototyping, feasibility studies for exploratory clinical research · Support the productization of research prototypes<p>We look for: · Strong research capability in computer vision, machine learning, text analytics and medical image analysis, proven by publications in journals/conferences. · Research experience in image/text analytics using large scale, weakly supervised / unsupervised learning algorithms · Research experience in medical image/text analysis of different modalities (CT, MRI, PET, medical reports etc.)<p>Email: Kevin.Brown@siemens-healthineers.com</p>
]]></description><pubDate>Fri, 03 Jan 2020 14:39:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=21946207</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=21946207</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=21946207</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Breast cancer detection in mammography using deep learning approach"]]></title><description><![CDATA[
<p>If you find this kind work interesting, our AI group at Siemens Healthineers is hiring interns to carry out projects like this. We typically target machine learning or medical imaging PhD students, but are open to a variety of backgrounds. Please feel free to reach out via email.</p>
]]></description><pubDate>Tue, 31 Dec 2019 00:52:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=21918111</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=21918111</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=21918111</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Ask HN: Who is hiring? (December 2019)"]]></title><description><![CDATA[
<p>Siemens Healthineers Imaging Intelligence | <a href="https://www.siemens-healthineers.com" rel="nofollow">https://www.siemens-healthineers.com</a> | Malvern, PA (Greater Philadelphia) | INTERNS | Onsite<p>Our R&D group delivers medical image/text tools (e.g. deep learning, NLP, etc) for medical data analysis. We are well recognized for delivering cutting-edge intelligent solutions to Siemens 3D workstations and medical imaging scanners. Our group also has strong publication record in top tier journals and conferences, and several Siemens "inventor of the year" award recipients.<p>We offer well-paid internships lasting >= 3 months, with independent moonshot projects.<p>Responsibilities:
·    Contribute to research projects to develop intelligent solutions for medical imaging and text analytics
·    Conduct fast prototyping, feasibility studies for exploratory clinical research
·    Support the productization of research prototypes<p>We look for:
·    Strong research capability in computer vision, machine learning, text analytics and medical image analysis, proven by publications in journals/conferences.
·    Research experience in image/text analytics using large scale, weakly supervised / unsupervised learning algorithms
·    Research experience in medical image/text analysis of different modalities (CT, MRI, PET, medical reports etc.)<p>Email: Kevin.Brown@siemens-healthineers.com</p>
]]></description><pubDate>Mon, 02 Dec 2019 19:25:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=21685866</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=21685866</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=21685866</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Nearly 400 medical devices, procedures and practices found ineffective in study"]]></title><description><![CDATA[
<p>I read the original study. I have a few thoughts. My partner is a physician and I am an AI researcher working in medicine. I think a lot about doctors as machine learning models, and RCT results like loss terms in a complicated objective function.<p>What is the best learning rate for updating physicians (our models) from the results of RCT's (part of our loss)?<p>The authors reviewed all articles in three journals from (generally) between 2003-2017. They didn't, afaict (please point me if they did), review the time-to-correction (if any correction has been made). It takes some time before the results of an RCT end up in established practice. I'm actually surprised it's so small in many cases.<p>It's not like there's a database where the results of every RCT are immediately updated and the physician model is retrained overnight on the new data.<p>Even if there were, imagine if the learning rate (so to speak) were so high that every discipline immediately changed their published best practices on the basis of a single RCT?<p>Here's a cautionary paragraph from one of the excellent reversal studies they use:<p><i>Several limitations of the study warrant discussion. First, because we enrolled only 26% of eligible patients, our findings must be generalized cautiously. The most frequent reason that patients declined enrollment was a strong preference for one treatment or the other. Since patients' preferences may be associated with treatment outcome, our trial may be vulnerable to selection bias. Participating surgeons may not have referred potentially eligible patients because they were uncomfortable randomly assigning these patients to treatment; this form of selective enrollment may also create bias.26 Second, because the trial was conducted in academic referral centers, the findings should be generalized carefully to community settings. Third, we did not formally assess the fidelity of the physical therapists or surgeons to the standard intervention protocols. Finally, our study was not blinded, since our investigative group did not consider a sham comparison group feasible.</i> [0]<p>I'm less concerned about RCT to Best Practice time than from Best Practice to Typical Physician Practice time. There is a cascaded model connected to the 'complex RCT loss' and it's discipline published practice down to individual physician treating patients. Compressing the time from RCT to individual physician is fraught with difficulties, but could be improved.<p>Finally, RCT is the gold standard, but it's not perfect and it doesn't always clearly translate to the individual physician's model of practice. Many best practices weren't established from RCT's either.<p>And an inconclusive result from an RCT is not the same thing as proving that there's no difference in outcomes, but a proper statistician can chime in there.<p>[0]<a href="https://www.nejm.org/doi/full/10.1056/NEJMoa1301408" rel="nofollow">https://www.nejm.org/doi/full/10.1056/NEJMoa1301408</a></p>
]]></description><pubDate>Thu, 13 Jun 2019 15:48:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=20175998</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=20175998</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=20175998</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "A former lead designer of Gmail fixes Gmail with a Chrome extension"]]></title><description><![CDATA[
<p>It's funny, I was thinking that Inbox did this nicely, and of course he did that too.</p>
]]></description><pubDate>Fri, 26 Apr 2019 03:06:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=19754508</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=19754508</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=19754508</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Opinion: A.I. Could Worsen Health Disparities"]]></title><description><![CDATA[
<p>Some medical devices also have other net cost benefits though. For example, shorter, less intensive hospital stays from less invasive testing.</p>
]]></description><pubDate>Thu, 31 Jan 2019 15:13:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=19045234</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=19045234</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=19045234</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Opinion: A.I. Could Worsen Health Disparities"]]></title><description><![CDATA[
<p>Dr Khullar suggests that AI will exacerbate biases in medical practice. His fundamental concern is that machine learning will codify biases and become self-fulfilling prophesies. But there is scant evidence that AI will <i>worsen</i> these disparities.<p>If anything, a machine-learning point of view better addresses his concerns than a traditional one, because they can be much more quickly updated to correct for identified biases. Doctors spend years and years of hard work becoming efficient and effective human algorithms themselves, and updating those human algorithms in the face of newer evidence is difficult. In standard practice, biases are often invisible and uncodified to begin with. "Moral intuition" is something all doctors use, but it's also something of a black box in nearly every real-world use case.</p>
]]></description><pubDate>Thu, 31 Jan 2019 14:21:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=19044829</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=19044829</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=19044829</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Advanced Data Analysis from an Elementary Point of View (2017) [pdf]"]]></title><description><![CDATA[
<p>His writing is somewhat tongue in cheek.<p>For whatever it’s worth, he seems to be a dedicated teacher who posts self criticisms of his courses publicly online. The course this book is based on has grown quite successful as well.<p>I’m not sure if you’re just trolling, but encouraging others to avoid this book might be a mistake. It’s quite good.</p>
]]></description><pubDate>Mon, 19 Feb 2018 15:29:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=16413417</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=16413417</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=16413417</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "Some Insights from a Julia Developer"]]></title><description><![CDATA[
<p>The packages-first attitude feels significant to me. A language that “users” enjoy but package developers also enjoy seems important. I hadn’t thought about language choice from a heavily package-development weighted perspective before. It seems obvious in retrospect though, which is probably a sign of something cool.<p>A version of this would be: how can good package development be as easy as possible, and how can package use be as easy as possible?<p>I haven’t done any serious work in Julia mainly because the python libraries are mature, good, and performant enough. I can’t speak for everyone, but for end users in science labs library support is perhaps the biggest consideration for language choice.</p>
]]></description><pubDate>Sat, 14 Oct 2017 14:10:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=15472233</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=15472233</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=15472233</guid></item><item><title><![CDATA[New comment by kevinalexbrown in "The Asynchronous Computability Theorem"]]></title><description><![CDATA[
<p>This is an important point, but one I feel the article acknowledges well.<p>There are many impossibility problems that can be solved by relaxing some constraints (like wait-free) or by accepting some unsolvability (we don’t worry too hard about our programs halting in practice, and we generally trust compressed sensing results because the probability of failure is provably delta, say), or just by changing some desiderata.<p>Great examples include arrows impossibility theorem or the no good clustering result, each of which have solutions if the assumptions that operationalize our intuition are changed slightly.</p>
]]></description><pubDate>Tue, 10 Oct 2017 03:17:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=15439050</link><dc:creator>kevinalexbrown</dc:creator><comments>https://news.ycombinator.com/item?id=15439050</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=15439050</guid></item></channel></rss>