<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: tomsanbear</title><link>https://news.ycombinator.com/user?id=tomsanbear</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 03 Aug 2026 23:26:34 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=tomsanbear" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by tomsanbear in "Andy Pavlo joins ClickHouse to establish ClickHouse Labs"]]></title><description><![CDATA[
<p>Always enjoyed his lecture series from CMU, hopefully those continue in a sponsored format from Clickhouse.</p>
]]></description><pubDate>Mon, 03 Aug 2026 15:28:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=49157078</link><dc:creator>tomsanbear</dc:creator><comments>https://news.ycombinator.com/item?id=49157078</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49157078</guid></item><item><title><![CDATA[New comment by tomsanbear in "Claude Sonnet 4 now supports 1M tokens of context"]]></title><description><![CDATA[
<p>I just want a better way to invalidate old context... It's great that I can fit more context, but the main challenge is claude getting sidetracked with 10 invalid grep calls, pytest dumping a 10k token stack trace etc.... And yes the ability to go back in time via esc+esc is great but I want claude to read the error stack learn from it and purge from its context or at least let me lobotomize ot selectively... 
Learning and discarding the raw output from tool calls feels like the missing piece here still.</p>
]]></description><pubDate>Wed, 13 Aug 2025 00:42:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=44883426</link><dc:creator>tomsanbear</dc:creator><comments>https://news.ycombinator.com/item?id=44883426</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44883426</guid></item></channel></rss>