<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: GeoffNN</title><link>https://news.ycombinator.com/user?id=GeoffNN</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Tue, 20 Oct 2026 03:09:25 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=GeoffNN" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by GeoffNN in "Timestral: Mistral release time-series foundation models"]]></title><description><![CDATA[
<p>Not that I know of!</p>
]]></description><pubDate>Mon, 19 Oct 2026 13:09:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=50152001</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=50152001</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50152001</guid></item><item><title><![CDATA[New comment by GeoffNN in "Timestral: Mistral release time-series foundation models"]]></title><description><![CDATA[
<p>omg - to keep the record straight, this person is not affiliated with The Forecasting Company</p>
]]></description><pubDate>Mon, 19 Oct 2026 13:03:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=50151948</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=50151948</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50151948</guid></item><item><title><![CDATA[T0: A Time-Series Foundation Model for Forecasting with Context]]></title><description><![CDATA[
<p>Article URL: <a href="https://arxiv.org/abs/2609.24559">https://arxiv.org/abs/2609.24559</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50151913">https://news.ycombinator.com/item?id=50151913</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 19 Oct 2026 12:59:43 +0000</pubDate><link>https://arxiv.org/abs/2609.24559</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=50151913</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50151913</guid></item><item><title><![CDATA[New comment by GeoffNN in "Timestral: Mistral release time-series foundation models"]]></title><description><![CDATA[
<p>Lots of friendly connections :) Both in Paris!
Also same shape of model that we know well</p>
]]></description><pubDate>Mon, 19 Oct 2026 12:56:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=50151861</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=50151861</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50151861</guid></item><item><title><![CDATA[New comment by GeoffNN in "Timestral: Mistral release time-series foundation models"]]></title><description><![CDATA[
<p>Not all of their trading is high freq</p>
]]></description><pubDate>Mon, 19 Oct 2026 12:46:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=50151760</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=50151760</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50151760</guid></item><item><title><![CDATA[New comment by GeoffNN in "Timestral: Mistral release time-series foundation models"]]></title><description><![CDATA[
<p>Not dunking! Adding info/context for this field. We're very excited to see Mistral is building here too!</p>
]]></description><pubDate>Mon, 19 Oct 2026 12:46:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=50151754</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=50151754</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50151754</guid></item><item><title><![CDATA[New comment by GeoffNN in "Timestral: Mistral's time-series foundation models"]]></title><description><![CDATA[
<p>Ah thanks!
The Mistral model has a restrictive license, where you need a specific commercial agreement to use it if you make more than 20M$ revenue/month. Ours is Apache 2.<p>If you compare the 2, it's not a strict "timestral is better" story - the winrates are tight - your probably should try both on your use-case. We have a 42% winrate vs Timestral on fevbench for instance.<p>Ofc, our main value is not just these models, but also 1) extra capabilities (like our model that can natively account for stockouts/censored data, and others coming soon), 2) our systems and products around the model, and 3) we're all laser focused on time-series/forecasting</p>
]]></description><pubDate>Mon, 19 Oct 2026 12:43:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=50151727</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=50151727</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50151727</guid></item><item><title><![CDATA[New comment by GeoffNN in "Show HN: machine0 – Persistent NixOS VMs You Control from the CLI"]]></title><description><![CDATA[
<p>Very cool! I'll definitely check this out for auto-research like experiments, esp. with GPUs</p>
]]></description><pubDate>Mon, 15 Jun 2026 20:57:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=48546947</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=48546947</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48546947</guid></item><item><title><![CDATA[New comment by GeoffNN in "A decoder-only foundation model for time-series forecasting"]]></title><description><![CDATA[
<p>They say they can optionally use some time-dependent covariates (derived from the datetime) -- did anyone figure out if/how they are actually used in this model from the paper?</p>
]]></description><pubDate>Mon, 05 Feb 2024 15:07:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=39261965</link><dc:creator>GeoffNN</dc:creator><comments>https://news.ycombinator.com/item?id=39261965</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39261965</guid></item></channel></rss>