<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: tristanz</title><link>https://news.ycombinator.com/user?id=tristanz</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Tue, 16 Jun 2026 05:58:53 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=tristanz" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[A Peek into Tesla's Autonomous Future by VP Ashok Elluswamy at ICCV25 [video]]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.youtube.com/watch?v=IRu-cPkpiFk">https://www.youtube.com/watch?v=IRu-cPkpiFk</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=45908188">https://news.ycombinator.com/item?id=45908188</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 12 Nov 2025 23:08:00 +0000</pubDate><link>https://www.youtube.com/watch?v=IRu-cPkpiFk</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=45908188</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45908188</guid></item><item><title><![CDATA[New comment by tristanz in "Tools: Code Is All You Need"]]></title><description><![CDATA[
<p>Yes, I think you could get quite far with a few tools like memory/todo list + code interpreter + script save/load.  You could probably get a lot farther though if you RLVRed this similar to how o3 uses web search so effectively during it's thinking process.</p>
]]></description><pubDate>Thu, 03 Jul 2025 17:04:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=44457101</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=44457101</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44457101</guid></item><item><title><![CDATA[New comment by tristanz in "Tools: Code Is All You Need"]]></title><description><![CDATA[
<p>You can combine MCPs within composable LLM generated code if you put in a little work. At Continual (<a href="https://continual.ai" rel="nofollow">https://continual.ai</a>), we have many workflows that require bulk actions, e.g. iterating over all issues, files, customers, etc.  We inject MCP tools into a sandboxed code interpreter and have the agent generate both direct MCP tool calls and composable scripts that leverage MCP tools depending on the task complexity.  After a bunch of work it actually works quite well.  We are also experimenting with continual learning via a Voyager like approach where the LLM can save tool scripts for future use, allowing lifelong learning for repeated workflows.</p>
]]></description><pubDate>Thu, 03 Jul 2025 15:55:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=44456336</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=44456336</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44456336</guid></item><item><title><![CDATA[New comment by tristanz in "The Problem with AI Welfare"]]></title><description><![CDATA[
<p>Not considering the potential for AI consciousness and suffering seems very shortsighted.  There are plausible reasons to believe that both could emerge from an RL processes coupled with small architectural and data regime changes.  Today's models have inherent architectural limits around continual learning that make this unlikely, but that will change.</p>
]]></description><pubDate>Thu, 12 Jun 2025 14:37:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=44258327</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=44258327</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44258327</guid></item><item><title><![CDATA[New comment by tristanz in "ORPC 1.0 – Typesafe APIs made simple"]]></title><description><![CDATA[
<p>Coming from TRPC, ORPC is a breath of fresh air.  I tested out a migration from TRPC to ORPC and I found two things particularly beneficial.<p>The first is contract-first development which separates the contract from the implementation.  This allows you to avoid codebase dependencies between your server and client.  TRPC works fine when you only use your client from your server package, but if you need to export it elsewhere, e.g. a public SDK, you can easily end up with circular dependency issues and a bunch of pain.<p>The second is OpenAPI support.  TRPC doesn't support OpenAPI generation and trpc-openapi is unmaintained.  ORPC has first class OpenAPI support, which means you can use ORPC internally but expose a public OpenAPI API to customers and generate OpenAPI based clients if you want to.<p>I'm hoping this project gets traction since it is amazingly well done.  I have zero affiliation or interest in ORPC to be clear, I just loved it from my quick tests.</p>
]]></description><pubDate>Tue, 15 Apr 2025 16:51:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=43695451</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=43695451</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43695451</guid></item><item><title><![CDATA[ORPC 1.0 – Typesafe APIs made simple]]></title><description><![CDATA[
<p>Article URL: <a href="https://orpc.unnoq.com/blog/v1-announcement">https://orpc.unnoq.com/blog/v1-announcement</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=43695365">https://news.ycombinator.com/item?id=43695365</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Tue, 15 Apr 2025 16:44:28 +0000</pubDate><link>https://orpc.unnoq.com/blog/v1-announcement</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=43695365</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43695365</guid></item><item><title><![CDATA[New comment by tristanz in "OpenAI: Streaming is now available in the Assistants API"]]></title><description><![CDATA[
<p>As an FYI, this is fine for rough usage, but it's not accurate.  The OpenAI APIs inject various tokens you are unaware of into the input for things like function calling.</p>
]]></description><pubDate>Thu, 14 Mar 2024 04:15:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=39700553</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=39700553</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39700553</guid></item><item><title><![CDATA[The AI Revenue Wave]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.continual.ai/blog-post/the-ai-revenue-wave">https://www.continual.ai/blog-post/the-ai-revenue-wave</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=38492406">https://news.ycombinator.com/item?id=38492406</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Fri, 01 Dec 2023 21:06:57 +0000</pubDate><link>https://www.continual.ai/blog-post/the-ai-revenue-wave</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=38492406</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=38492406</guid></item><item><title><![CDATA[Drive Anywhere: End-to-End Autonomous Driving with Multimodal Foundation Models]]></title><description><![CDATA[
<p>Article URL: <a href="https://drive-anywhere.github.io/">https://drive-anywhere.github.io/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=38060660">https://news.ycombinator.com/item?id=38060660</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 29 Oct 2023 17:19:44 +0000</pubDate><link>https://drive-anywhere.github.io/</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=38060660</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=38060660</guid></item><item><title><![CDATA[Black holes as the source of dark energy]]></title><description><![CDATA[
<p>Article URL: <a href="https://aasnova.org/2023/02/15/black-holes-as-the-source-of-dark-energy/">https://aasnova.org/2023/02/15/black-holes-as-the-source-of-dark-energy/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=34839090">https://news.ycombinator.com/item?id=34839090</a></p>
<p>Points: 113</p>
<p># Comments: 77</p>
]]></description><pubDate>Fri, 17 Feb 2023 19:23:46 +0000</pubDate><link>https://aasnova.org/2023/02/15/black-holes-as-the-source-of-dark-energy/</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=34839090</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=34839090</guid></item><item><title><![CDATA[It's not you, it's SQL]]></title><description><![CDATA[
<p>Article URL: <a href="https://stack.convex.dev/not-sql">https://stack.convex.dev/not-sql</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=34582549">https://news.ycombinator.com/item?id=34582549</a></p>
<p>Points: 61</p>
<p># Comments: 83</p>
]]></description><pubDate>Mon, 30 Jan 2023 16:54:04 +0000</pubDate><link>https://stack.convex.dev/not-sql</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=34582549</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=34582549</guid></item><item><title><![CDATA[New comment by tristanz in "ChatGPT won’t replace search engines any time soon"]]></title><description><![CDATA[
<p>ChatGTP maybe, but OpenAI hasn't even tried to train a model to replace search.<p>Until somebody tries to fine-tune a model using RLHF explicitly with the goal of replacing Google it's very hard to know what the resulting experience would look like.  It could be shocking if ChatGTP is any guide.</p>
]]></description><pubDate>Sun, 08 Jan 2023 01:20:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=34294989</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=34294989</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=34294989</guid></item><item><title><![CDATA[New comment by tristanz in "“Twitter will be forming a content moderation council with diverse viewpoints”"]]></title><description><![CDATA[
<p><a href="https://www.zdnet.com/article/what-is-elon-musks-x-the-everything-app/" rel="nofollow">https://www.zdnet.com/article/what-is-elon-musks-x-the-every...</a></p>
]]></description><pubDate>Fri, 28 Oct 2022 21:03:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=33377889</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=33377889</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=33377889</guid></item><item><title><![CDATA[Dylan Field pitches and demos seed-stage Figma]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.youtube.com/watch?v=C1UUVdN3kdQ">https://www.youtube.com/watch?v=C1UUVdN3kdQ</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=32855171">https://news.ycombinator.com/item?id=32855171</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 15 Sep 2022 17:09:18 +0000</pubDate><link>https://www.youtube.com/watch?v=C1UUVdN3kdQ</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=32855171</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=32855171</guid></item><item><title><![CDATA[New comment by tristanz in "PRQL – A proposal for a better SQL"]]></title><description><![CDATA[
<p>To get 10x I think you need to wed it to solving broader workflow challenges, like dbt does today.</p>
]]></description><pubDate>Mon, 24 Jan 2022 22:35:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=30065021</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=30065021</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=30065021</guid></item><item><title><![CDATA[The Self Provisioning Runtime]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.swyx.io/self-provisioning-runtime/">https://www.swyx.io/self-provisioning-runtime/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=29754608">https://news.ycombinator.com/item?id=29754608</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Sat, 01 Jan 2022 01:48:15 +0000</pubDate><link>https://www.swyx.io/self-provisioning-runtime/</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=29754608</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29754608</guid></item><item><title><![CDATA[New comment by tristanz in "Build models like we build open-source software"]]></title><description><![CDATA[
<p>Collaborative incremental improvement of models would be extremely disruptive.  While this happens via research, it's massively inefficient, particularly as pretrained models get larger and span multiple modalities.</p>
]]></description><pubDate>Wed, 08 Dec 2021 17:01:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=29487238</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=29487238</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29487238</guid></item><item><title><![CDATA[Build models like we build open-source software]]></title><description><![CDATA[
<p>Article URL: <a href="https://colinraffel.com/blog/a-call-to-build-models-like-we-build-open-source-software.html">https://colinraffel.com/blog/a-call-to-build-models-like-we-build-open-source-software.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=29487212">https://news.ycombinator.com/item?id=29487212</a></p>
<p>Points: 51</p>
<p># Comments: 8</p>
]]></description><pubDate>Wed, 08 Dec 2021 16:59:26 +0000</pubDate><link>https://colinraffel.com/blog/a-call-to-build-models-like-we-build-open-source-software.html</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=29487212</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29487212</guid></item><item><title><![CDATA[The Modern Data Stack Ecosystem – Fall 2021 Edition]]></title><description><![CDATA[
<p>Article URL: <a href="https://continual.ai/post/the-modern-data-stack-ecosystem-fall-2021-edition">https://continual.ai/post/the-modern-data-stack-ecosystem-fall-2021-edition</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=29033931">https://news.ycombinator.com/item?id=29033931</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Fri, 29 Oct 2021 00:42:00 +0000</pubDate><link>https://continual.ai/post/the-modern-data-stack-ecosystem-fall-2021-edition</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=29033931</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29033931</guid></item><item><title><![CDATA[New comment by tristanz in "Ask HN: Who is hiring? (September 2021)"]]></title><description><![CDATA[
<p>Continual | Frontend, Fullstack, and ML Engineer | SF, REMOTE | <a href="https://continual.ai" rel="nofollow">https://continual.ai</a>.<p>Continual is building the missing AI layer for the modern data stack.  We're hiring multiple roles (frontend, full stack, ML engineer) at all levels.  This is an opportunity to get in early on a massive opportunity in one of the most interesting areas of technology – democratizing operational AI/ML.<p>Email tristan@continual.ai to learn more.</p>
]]></description><pubDate>Wed, 01 Sep 2021 16:27:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=28381955</link><dc:creator>tristanz</dc:creator><comments>https://news.ycombinator.com/item?id=28381955</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=28381955</guid></item></channel></rss>