<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: jamesbrady</title><link>https://news.ycombinator.com/user?id=jamesbrady</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 19 Nov 2026 21:44:05 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=jamesbrady" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[What distinguishes great engineers when AI writes the code?]]></title><description><![CDATA[
<p>Article URL: <a href="https://elicit.com/blog/engineering-interviews-in-the-era-of-agents">https://elicit.com/blog/engineering-interviews-in-the-era-of-agents</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47618548">https://news.ycombinator.com/item?id=47618548</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 02 Apr 2026 18:46:59 +0000</pubDate><link>https://elicit.com/blog/engineering-interviews-in-the-era-of-agents</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=47618548</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47618548</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (November 2025)"]]></title><description><![CDATA[
<p>Elicit | San Francisco + remote (US timezones) | <a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a><p>Elicit is building AI for science. We scaled to >$7MM annual revenue and 450k MAU fast. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.<p>Reasons you might want to join Elicit:<p>1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (<a href="https://ought.org/" rel="nofollow">https://ought.org/</a>) which was an applied research lab.<p>2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. <a href="https://ought.org/updates/2022-04-06-process" rel="nofollow">https://ought.org/updates/2022-04-06-process</a> for a primer.<p>3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.<p>Reasons you might not want to join Elicit:<p>1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.<p>2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.<p>A selection of open roles—more on the site:<p>• ML engineer: <a href="https://elicit.com/careers/913a03d5-bd26-4c64-8346-21a029f344f7?utm_source=hn" rel="nofollow">https://elicit.com/careers/913a03d5-bd26-4c64-8346-21a029f34...</a><p>• Founding data engineer: <a href="https://elicit.com/careers/8cc10be7-30ca-469a-a567-dfcdb271f149?utm_source=hn" rel="nofollow">https://elicit.com/careers/8cc10be7-30ca-469a-a567-dfcdb271f...</a><p>• Senior software engineer: <a href="https://elicit.com/careers/aa99e2e9-5b15-4cd3-ac9d-9c9177ff61c8?utm_source=hn" rel="nofollow">https://elicit.com/careers/aa99e2e9-5b15-4cd3-ac9d-9c9177ff6...</a></p>
]]></description><pubDate>Mon, 03 Nov 2025 18:56:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=45802840</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=45802840</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45802840</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (July 2025)"]]></title><description><![CDATA[
<p>Elicit | San Francisco + remote (US timezones) | <a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a><p>Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We scaled to >$6MM annual revenue and 450k MAU <i>fast</i>. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.<p>Reasons you might want to join Elicit:<p>1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (<a href="https://ought.org/" rel="nofollow">https://ought.org/</a>) which was an applied research lab.<p>2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. <a href="https://ought.org/updates/2022-04-06-process" rel="nofollow">https://ought.org/updates/2022-04-06-process</a> for a primer.<p>3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.<p>Reasons you might <i>not</i> want to join Elicit:<p>1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.<p>2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.<p>A selection of open roles—more on the site:<p>• Machine learning engineer: <a href="https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346-21a029f344f7" rel="nofollow">https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...</a><p>• Our first data engineer: <a href="https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753-2915ce9c04a7" rel="nofollow">https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...</a><p>• Front-end engineer: <a href="https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026-1fe2fe02c3eb" rel="nofollow">https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...</a></p>
]]></description><pubDate>Tue, 01 Jul 2025 18:46:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=44436910</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=44436910</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44436910</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (June 2025)"]]></title><description><![CDATA[
<p>Elicit | San Francisco + remote (US timezones) | <a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a><p>Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We scaled to >$6MM annual revenue and 450k MAU <i>fast</i>. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.<p>Reasons you might want to join Elicit:<p>1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (<a href="https://ought.org/" rel="nofollow">https://ought.org/</a>) which was an applied research lab.<p>2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. <a href="https://ought.org/updates/2022-04-06-process" rel="nofollow">https://ought.org/updates/2022-04-06-process</a> for a primer.<p>3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.<p>Reasons you might <i>not</i> want to join Elicit:<p>1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.<p>2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.<p>A selection of open roles—more on the site:<p>• Machine learning engineer: <a href="https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346-21a029f344f7" rel="nofollow">https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...</a><p>• Our first data engineer: <a href="https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753-2915ce9c04a7" rel="nofollow">https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...</a><p>• Front-end engineer: <a href="https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026-1fe2fe02c3eb" rel="nofollow">https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...</a></p>
]]></description><pubDate>Mon, 02 Jun 2025 16:55:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=44160799</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=44160799</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44160799</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (April 2025)"]]></title><description><![CDATA[
<p>Elicit | San Francisco + remote (US timezones) | <a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a><p>Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We scaled to >$5MM annual revenue and 450k MAU <i>fast</i>. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.<p>Reasons you might want to join Elicit:<p>1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (<a href="https://ought.org/" rel="nofollow">https://ought.org/</a>) which was an applied research lab.<p>2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. <a href="https://ought.org/updates/2022-04-06-process" rel="nofollow">https://ought.org/updates/2022-04-06-process</a> for a primer.<p>3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.<p>Reasons you might <i>not</i> want to join Elicit:<p>1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.<p>2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.<p>A selection of open roles—more on the site:<p>• Front-end engineer: <a href="https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026-1fe2fe02c3eb" rel="nofollow">https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...</a><p>• Machine learning engineer: <a href="https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346-21a029f344f7" rel="nofollow">https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...</a><p>• Data engineer: <a href="https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753-2915ce9c04a7" rel="nofollow">https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...</a></p>
]]></description><pubDate>Tue, 01 Apr 2025 15:14:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=43547793</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=43547793</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43547793</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (March 2025)"]]></title><description><![CDATA[
<p>Elicit | San Francisco + remote (US timezones) | <a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a><p>Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We've scaled to >$3MM annual revenue and 400k MAU with our small team. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.<p>Reasons you might want to join Elicit:<p>1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (<a href="https://ought.org/" rel="nofollow">https://ought.org/</a>) which was more of a research lab.<p>2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. <a href="https://ought.org/updates/2022-04-06-process" rel="nofollow">https://ought.org/updates/2022-04-06-process</a> for a primer.<p>3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.<p>Reasons you might <i>not</i> want to join Elicit:<p>1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.<p>2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.<p>3. We just raised a Series A, so if you want existential dread (and concomitant founding-team-upside), Elicit is not for you.<p>A selection of open roles—more on the site:<p>• Front-end engineer: <a href="https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026-1fe2fe02c3eb" rel="nofollow">https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...</a><p>• Machine learning engineer: <a href="https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346-21a029f344f7" rel="nofollow">https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...</a><p>• Data engineer: <a href="https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753-2915ce9c04a7" rel="nofollow">https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...</a></p>
]]></description><pubDate>Mon, 03 Mar 2025 19:30:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=43245713</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=43245713</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43245713</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (February 2025)"]]></title><description><![CDATA[
<p>Elicit | San Francisco + remote (US timezones) | <a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a><p>Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We've scaled to >$3MM annual revenue and 400k MAU with our small team. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.<p>Reasons you might want to join Elicit:<p>1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (<a href="https://ought.org/" rel="nofollow">https://ought.org/</a>) which was more of a research lab.<p>2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. <a href="https://ought.org/updates/2022-04-06-process" rel="nofollow">https://ought.org/updates/2022-04-06-process</a> for a primer.<p>3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.<p>Reasons you might <i>not</i> want to join Elicit:<p>1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.<p>2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.<p>A selection of open roles—more on the site:<p>• Front-end engineer: <a href="https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026-1fe2fe02c3eb" rel="nofollow">https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...</a><p>• Machine learning engineer: <a href="https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346-21a029f344f7" rel="nofollow">https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...</a><p>• Data engineer: <a href="https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753-2915ce9c04a7" rel="nofollow">https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...</a></p>
]]></description><pubDate>Mon, 03 Feb 2025 16:38:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=42920024</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=42920024</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42920024</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (January 2025)"]]></title><description><![CDATA[
<p>Elicit | San Francisco + remote (US timezones) | <a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a><p>Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We've scaled to >$3MM annual revenue and 400k MAU with our small team. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.<p>Reasons you might want to join Elicit:<p>1. We have been thinking about how to apply AI for high-stakes knowledge work since even before GPT-1. Elicit was spun-out from Ought (<a href="https://ought.org/" rel="nofollow">https://ought.org/</a>) which was more of a research lab.<p>2. If you're concerned about the potential downsides and risks of super-powerful AI: so are we! Happy to explain more of this story on a call, or you can also check e.g. <a href="https://ought.org/updates/2022-04-06-process" rel="nofollow">https://ought.org/updates/2022-04-06-process</a> for a primer.<p>3. We like spending time with each other, and have get-togethers every 6 weeks, but we're flexible on location.<p>Reasons you might <i>not</i> want to join Elicit:<p>1. We've found a level of product market fit (as evidenced by metrics) but we've not proven everything out yet. If you want predictability, Elicit is not for you.<p>2. We aim to hire people who thrive in low-bureaucracy environments, and let them loose. If you need structure, lots of support, close mentorship, or tightly-defined work assignments, Elicit is not for you.<p>A selection of open roles—more on the site:<p>• Front-end engineer: <a href="https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026-1fe2fe02c3eb" rel="nofollow">https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...</a><p>• Machine learning engineer: <a href="https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346-21a029f344f7" rel="nofollow">https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...</a><p>• Data engineer: <a href="https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753-2915ce9c04a7" rel="nofollow">https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...</a></p>
]]></description><pubDate>Thu, 02 Jan 2025 16:57:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=42576133</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=42576133</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42576133</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (December 2024)"]]></title><description><![CDATA[
<p>Elicit | San Francisco + remote (US timezones) | <a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a><p>Elicit is an AI research assistant that uses language models to help researchers figure out what’s true and make better decisions. We've scaled to >$2MM annual revenue and 400k MAU with our small team. We're hiring for multiple roles and are primarily interested in people with early-stage company experience, who are comfortable in high-agency, fast-paced teams.<p>Front-end engineer: <a href="https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026-1fe2fe02c3eb" rel="nofollow">https://elicit.com/careers?ashby_jid=b5e218b8-8730-4254-b026...</a><p>Machine learning engineer: <a href="https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346-21a029f344f7" rel="nofollow">https://elicit.com/careers?ashby_jid=913a03d5-bd26-4c64-8346...</a><p>Data engineer: <a href="https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753-2915ce9c04a7" rel="nofollow">https://elicit.com/careers?ashby_jid=4617f630-f971-4716-b753...</a></p>
]]></description><pubDate>Wed, 04 Dec 2024 13:51:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=42317406</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=42317406</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42317406</guid></item><item><title><![CDATA[New comment by jamesbrady in "Ask HN: Who is hiring? (July 2024)"]]></title><description><![CDATA[
<p>Elicit (<a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a>) | Oakland, CA and Hybrid | Frontend, AI, and Full-stack software engineering roles.<p>Elicit is automating high-quality reasoning so that we can help the world make more breakthroughs in every domain: from climate change to the gut microbiome to longevity and economic policy.<p>We’ve scaled to over 200,000 monthly users purely by word of mouth and recently crossed $1.5MM in annual revenue, 7 months after launching subscriptions.<p>We’re now building out our software engineering team, and hiring across several technical roles.<p>If you'd like to know more about some of the work we're doing you could check:
- A recent blog post current UX work: <a href="https://blog.elicit.com/living-documents-ai-ux/" rel="nofollow">https://blog.elicit.com/living-documents-ai-ux/</a>
- Me talking about our "AI engineer" role on a podcast a couple of weeks ago: <a href="https://www.latent.space/p/hiring" rel="nofollow">https://www.latent.space/p/hiring</a><p>Check out the careers page linked above, or email me at james@elicit.com</p>
]]></description><pubDate>Mon, 01 Jul 2024 20:32:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=40850326</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=40850326</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40850326</guid></item><item><title><![CDATA[New comment by jamesbrady in "Elicit – AI Research Assistant"]]></title><description><![CDATA[
<p>Elician here!<p>Our main focus is a little different to SciSummary actually. We're focussed on understanding researchers broader workflows, and providing a research assistant (i.e. rather than a particular narrow tool for summarisation or search).<p>The workflows we're most excited about at the moment are literature and systematic reviews: we think we can make these orders of magnitude faster and higher quality.</p>
]]></description><pubDate>Thu, 16 May 2024 14:47:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=40379113</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=40379113</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40379113</guid></item><item><title><![CDATA[New comment by jamesbrady in "Elicit – AI Research Assistant"]]></title><description><![CDATA[
<p>Actually, it's not an LLM!<p>We do use LLMs, but the secret sauce is an approach we call Factored Cognition which we wrote about here: <a href="https://ought.org/research/factored-cognition" rel="nofollow">https://ought.org/research/factored-cognition</a><p>(Elicit the company and app was spun out from Ought the research lab).<p>We do joke internally about the homophone (in fact, IIRC we did a little joke on our CEO by rebranding for his birthday in 2022) but I'm sorry to report that we're all careful, ethical, and well-behaved people :(</p>
]]></description><pubDate>Thu, 16 May 2024 14:44:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=40379087</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=40379087</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40379087</guid></item><item><title><![CDATA[New comment by jamesbrady in "Elicit – AI Research Assistant"]]></title><description><![CDATA[
<p>Oh, my bad—I misunderstood, thanks for the clarification</p>
]]></description><pubDate>Thu, 16 May 2024 14:41:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=40379056</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=40379056</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40379056</guid></item><item><title><![CDATA[New comment by jamesbrady in "Elicit – AI Research Assistant"]]></title><description><![CDATA[
<p>Elician here: thank you for sharing our tool and for this praise!<p>We're glad you're enjoying it.</p>
]]></description><pubDate>Thu, 16 May 2024 14:40:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=40379041</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=40379041</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40379041</guid></item><item><title><![CDATA[New comment by jamesbrady in "Elicit – AI Research Assistant"]]></title><description><![CDATA[
<p>Elician here.<p>This is a good point! (Hopefully) obviously, if we knew a particular claim was fishy, we wouldn't make it in the app in the first place.<p>However, we do do a couple of things which go towards addressing your concern:<p>1. We can be more or less confident in the answers we're giving in the app, and if that confidence dips below a threshold we mark that particular cell in the results table with a red warning icon which encourages caution and user verification. This confidence level isn't perfectly calibrated, of course, but we are trying to engender a healthy, active, wariness in our users so that they don't take Elicit results as gospel.
2. We provide sources for all of the claims made in the app. You can see these by clicking on any cell in the results table. We encourage users to check—or at least spot-check—the results which they are updating on. This verification is generally <i>much</i> faster than doing the generation of the answer in the first place.</p>
]]></description><pubDate>Thu, 16 May 2024 14:38:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=40379025</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=40379025</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40379025</guid></item><item><title><![CDATA[New comment by jamesbrady in "Elicit – AI Research Assistant"]]></title><description><![CDATA[
<p>Elician here! Thanks for your comment.<p>I'm not sure I agree that those rule-of-thumb statistics are "arbitrary" or "fictional"… I guess it depends on what you mean by that. I can say that on our part they're a good faith attempt to help users calibrate how best to use the tool, using evaluations of Elicit based on real usage.<p>Definitely accept that the tool can work better or worse depending on your domain or workflow though!<p>One way we do try to distinguish ourselves from vanilla LLMs is that we provide sources for all of the claims made. I mention this because we hope our users can approach the falsification process you mention for Google. We want to show people where particular claims come from such that we earn their trust.<p>Walking citation trails and verifying transitive claims is something we've talked about but need more people to implement! (<a href="https://elicit.com/careers" rel="nofollow">https://elicit.com/careers</a>)</p>
]]></description><pubDate>Thu, 16 May 2024 14:30:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=40378943</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=40378943</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40378943</guid></item><item><title><![CDATA[New comment by jamesbrady in "Elicit – AI Research Assistant"]]></title><description><![CDATA[
<p>Elician here!<p>Accuracy and supportedness of the claims made in Elicit are two of the most central things we focus on—it's a shame it didn't work as well as we'd like in this case.<p>I'd appreciate knowing more about the specifics so we can understand and improve</p>
]]></description><pubDate>Thu, 16 May 2024 14:23:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=40378847</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=40378847</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40378847</guid></item><item><title><![CDATA[New comment by jamesbrady in "Elicit – LLMs to Accelerate Research"]]></title><description><![CDATA[
<p>Elician here.<p>The two main problems we're addressing now are:<p>1. Finding papers / claims / data across an academic literature which is ballooning in size.<p>2. Using these raw materials to to answer questions in a reliable manner.<p>#2 is where the bulk of the tricky ML work is, and where vanilla language models often fall short because of limited context windows and hallucination.<p>We're also working to expand Elicit to help academics with other parts of their research, like surfacing critiques, suggesting related prior work, brainstorming related research questions, identifying risks of bias, …</p>
]]></description><pubDate>Tue, 26 Sep 2023 07:11:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=37655557</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=37655557</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37655557</guid></item><item><title><![CDATA[New comment by jamesbrady in "The expanding dark forest and generative AI"]]></title><description><![CDATA[
<p>This is not true.</p>
]]></description><pubDate>Wed, 04 Jan 2023 14:44:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=34246028</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=34246028</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=34246028</guid></item><item><title><![CDATA[New comment by jamesbrady in "How to quit like a boss"]]></title><description><![CDATA[
<p>Agree, way too long!<p>I gave that much of a heads-up of my leaving (not a formal resignation, as it happens) for a few reasons including:<p>1. It's a C-level replacement, which takes _months_<p>2. I cared deeply (and still do) about my team and the company mission, and wanted them to succeed: "the more notice the better" seemed right to me (it wasn't)<p>3. My career plans were to move into a new field which would require various courses, lots of reading, conferences, etc. – although I knew I was leaving, I hadn't even started speaking to potential next employers or investors when I told my CEO I was leaving<p>Putting all of that together is how I ended up with the 10 month blunder, but the post is less <i>against</i> 10 month notice periods, and more <i>for</i> matching the notice period to your handover, whatever that would be.<p>I feel your pain on working out those 5 weeks you mention! Awkward for everyone involved.</p>
]]></description><pubDate>Fri, 07 Jan 2022 04:57:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=29835113</link><dc:creator>jamesbrady</dc:creator><comments>https://news.ycombinator.com/item?id=29835113</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29835113</guid></item></channel></rss>