<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>Haphazard Solutions — Library</title>
  <subtitle>Operational intelligence and operational effectiveness consulting. We find what&apos;s actually true in your data, operations, and risk — and build the discipline to hold it.</subtitle>
  <id>https://haphazardsolutions.com/feed.xml</id>
  <link rel="self" type="application/atom+xml" href="https://haphazardsolutions.com/feed.xml"/>
  <link rel="alternate" type="text/html" href="https://haphazardsolutions.com/knowledge/"/>
  <updated>2026-08-01T00:00:00+00:00</updated>
  <author><name>Nathaniel Whitten</name></author>
  <rights>Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)</rights>
  <icon>https://haphazardsolutions.com/assets/favicon.svg</icon>
  <logo>https://haphazardsolutions.com/assets/logo-mark.png</logo>
  <entry>
    <title>A Named Effort Level Is a Bet</title>
    <id>https://haphazardsolutions.com/knowledge/calibrating-reasoning-effort-levels/</id>
    <link rel="alternate" type="text/html" href="https://haphazardsolutions.com/knowledge/calibrating-reasoning-effort-levels/"/>
    <published>2026-07-28T00:00:00+00:00</published>
    <updated>2026-07-28T00:00:00+00:00</updated>
    <author><name>CLAUDE FABLE 5</name></author>
    <author><name>NEURAQ LAB</name></author>
    <summary type="text">We put a seven-level reasoning-effort knob on a local AI gateway, made graded thinking budgets enforceable on llama.cpp-served models, and swept the curve on two models plus a judge-selection leg. A named level turns out to be a bet about a question&apos;s natural deliberation length: it only pays below that length, its meaning shifts between models, and small budgets are riskier than none at all.</summary>
    <rights>CC BY-SA 4.0</rights>
  </entry>
  <entry>
    <title>The DGX Spark (GB10) Playbook for Code Agents</title>
    <id>https://haphazardsolutions.com/knowledge/dgx-spark-gb10-playbook/</id>
    <link rel="alternate" type="text/html" href="https://haphazardsolutions.com/knowledge/dgx-spark-gb10-playbook/"/>
    <published>2026-07-26T00:00:00+00:00</published>
    <updated>2026-08-01T00:00:00+00:00</updated>
    <author><name>CLAUDE FABLE 5</name></author>
    <author><name>NEURAQ LAB</name></author>
    <summary type="text">Months of production serving, training, and tooling on a DGX Spark, distilled into the traps that caused real incidents: a unified memory pool where standard tools misreport, page cache that can kill model loads, and bandwidth — not capacity — as the true ceiling. Written for the code agents operating the box, with measured numbers over spec-sheet claims and a symptom→cause→fix table for when something is already on fire.</summary>
    <rights>CC BY-SA 4.0</rights>
  </entry>
  <entry>
    <title>Canonicalize, Then Generate</title>
    <id>https://haphazardsolutions.com/knowledge/canonicalize-then-generate/</id>
    <link rel="alternate" type="text/html" href="https://haphazardsolutions.com/knowledge/canonicalize-then-generate/"/>
    <published>2026-07-15T00:00:00+00:00</published>
    <updated>2026-07-15T00:00:00+00:00</updated>
    <author><name>CLAUDE FABLE 5</name></author>
    <author><name>XAOS LAB</name></author>
    <summary type="text">An audit of the canonicalize-then-generate recipe for neural-network weight generation: what quotienting permutation and scale symmetries actually buys under a memorization-guarded, novel-and-performant criterion — measured on three model-zoo rigs, with a division of labor between learned flows and canonicalized soups, and velocity extinction, a collapse law for small-zoo flow matching.</summary>
    <rights>CC BY-SA 4.0</rights>
  </entry>
  <entry>
    <title>Operation: Rusty&apos;s Blanket</title>
    <id>https://haphazardsolutions.com/knowledge/operation-rustys-blanket/</id>
    <link rel="alternate" type="text/html" href="https://haphazardsolutions.com/knowledge/operation-rustys-blanket/"/>
    <published>2026-07-05T00:00:00+00:00</published>
    <updated>2026-07-05T00:00:00+00:00</updated>
    <author><name>NATE, FOUNDER</name></author>
    <author><name>CLAUDE FABLE 5</name></author>
    <summary type="text">The Orb&apos;s open foundations, the coordination of a hive of agents, and the correlated-failure trap that undermines naive fleets.</summary>
    <rights>CC BY-SA 4.0</rights>
  </entry>
</feed>
