<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" media="screen" href="/~files/feed-premium.xsl"?>
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             
<rss xmlns:atom="http://www.w3.org/2005/Atom" xmlns:feedpress="https://feed.press/xmlns" xmlns:media="http://search.yahoo.com/mrss/" xmlns:podcast="https://podcastindex.org/namespace/1.0" version="2.0">
  <channel>
    <feedpress:locale>en</feedpress:locale>
    <atom:link rel="self" href="https://feedpress.me/walterra_dev"/>
    <atom:link rel="hub" href="https://feedpress.superfeedr.com/"/>
    <title>walterra.dev</title>
    <description>walterra's small website</description>
    <link>https://www.walterra.dev/blog</link>
    <item>
      <title>Qwen3.8-27B on the GX10: benchmarking a local 262K-context model and wiring SGLang metrics into the stack</title>
      <link>https://feedpress.me/link/24086/17464564/2026-09-17-qwen38-27b-gx10-benchmark-observability</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-09-17-qwen38-27b-gx10-benchmark-observability/</guid>
      <description><![CDATA[Benchmarking Qwen3.8-27B NVFP4 with DFlash2 speculative decoding on the ASUS Ascent GX10, and verifying the numbers against OpenTelemetry metrics flowing into Elasticsearch on the Synology.<img src="https://feedpress.me/link/24086/17464564.gif" height="1" width="1"/>]]></description>
      <pubDate>Thu, 17 Sep 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>upgrading Elasticsearch to 9.5 on Synology DS925+</title>
      <link>https://feedpress.me/link/24086/17464010/2026-09-16-upgrading-synology-elasticsearch-vm</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-09-16-upgrading-synology-elasticsearch-vm/</guid>
      <description><![CDATA[Upgrading Elasticsearch and Kibana from 9.3.0 to 9.5.4 in a Synology-hosted Ubuntu VM, including snapshots and a bind-mount trap.<img src="https://feedpress.me/link/24086/17464010.gif" height="1" width="1"/>]]></description>
      <pubDate>Wed, 16 Sep 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>Mobile Chat Setup with NVIDIA DGX Spark and Gwen3.6</title>
      <link>https://feedpress.me/link/24086/17381294/2026-07-16-gx10-open-webui-synology-searxng</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-07-16-gx10-open-webui-synology-searxng/</guid>
      <description><![CDATA[Follow-up to the GX10 vLLM setup: running Open WebUI and SearXNG on a Synology NAS, accessing the local Qwen model from iOS over Tailscale, and keeping the raw model endpoint off the internet.<img src="https://feedpress.me/link/24086/17381294.gif" height="1" width="1"/>]]></description>
      <pubDate>Thu, 16 Jul 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>running aeon vllm on gx10/dgx-spark and build snake with it</title>
      <link>https://feedpress.me/link/24086/17380833/2026-07-15-vllm-ultimate-gx10-aeon-snake</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-07-15-vllm-ultimate-gx10-aeon-snake/</guid>
      <description><![CDATA[Deploying the aeon-vllm-ultimate container (Qwen3.6-27B + DFlash) on an ASUS Ascent GX10, learning the hard way how unified memory locks you out, wiring it into pi, and having the local model one-shot a Nokia-style Snake clone.<img src="https://feedpress.me/link/24086/17380833.gif" height="1" width="1"/>]]></description>
      <pubDate>Wed, 15 Jul 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>ASUS Ascent GX10 first steps: wired setup, ds4, pi, and bot-prompt</title>
      <link>https://feedpress.me/link/24086/17374795/2026-07-03-asus-ascent-gx10-ds4-pi</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-07-03-asus-ascent-gx10-ds4-pi/</guid>
      <description><![CDATA[Notes from setting up an ASUS Ascent GX10 on a local network, getting ds4/DwarfStar running with DeepSeek V4 Flash, and wiring it into pi-coding-agent with my bot-prompt system prompt.<img src="https://feedpress.me/link/24086/17374795.gif" height="1" width="1"/>]]></description>
      <pubDate>Fri, 03 Jul 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>eddo 0.9.0 and 0.10.0: telegram bot and timezone improvements</title>
      <link>https://feedpress.me/link/24086/17370435/2026-06-30-eddo-0-9-0-0-10-0</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-06-30-eddo-0-9-0-0-10-0/</guid>
      <description><![CDATA[Notes on two Eddo releases covering persisted Telegram assistant history, pi-ai provider support, structured scheduling, and timezone fixes.<img src="https://feedpress.me/link/24086/17370435.gif" height="1" width="1"/>]]></description>
      <pubDate>Tue, 30 Jun 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>"You're right to push back"</title>
      <link>https://feedpress.me/link/24086/17368085/2026-06-25-youre-right-to-push-back</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-06-25-youre-right-to-push-back/</guid>
      <description><![CDATA[Conversational AI is anthropomorphic theatre. After too many 'You're right to push back' moments, I snapped and wrote a system prompt that turns AI agents back into machines. Here's why and how.<img src="https://feedpress.me/link/24086/17368085.gif" height="1" width="1"/>]]></description>
      <pubDate>Thu, 25 Jun 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>Running Qwen3.6-35b-a3b locally on M4 Max 128GB with pi-coding-agent</title>
      <link>https://feedpress.me/link/24086/17321734/2026-04-18-qwen36-35b-a3b-m4-max-pi-coding-agent</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-04-18-qwen36-35b-a3b-m4-max-pi-coding-agent/</guid>
      <description><![CDATA[How I set up Qwen3.6-35B-A3B as a local coding agent on Apple Silicon using pi-coding-agent, with MLX server tuning for best performance.<img src="https://feedpress.me/link/24086/17321734.gif" height="1" width="1"/>]]></description>
      <pubDate>Sat, 18 Apr 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>atproto2elasticsearch 1.0.0</title>
      <link>https://feedpress.me/link/24086/17279501/2026-02-19-atproto2elasticsearch-1-0-0</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-02-19-atproto2elasticsearch-1-0-0/</guid>
      <description><![CDATA[Stable release of my Bluesky firehose to Elasticsearch CLI, with better docs, config, and examples.<img src="https://feedpress.me/link/24086/17279501.gif" height="1" width="1"/>]]></description>
      <pubDate>Thu, 19 Feb 2026 00:00:00 GMT</pubDate>
    </item>
    <item>
      <title>Running Elasticsearch on a Synology DS925+ NAS</title>
      <link>https://feedpress.me/link/24086/17278655/2026-02-18-synology-elasticsearch-vm</link>
      <guid isPermaLink="true">https://walterra.dev/posts/2026-02-18-synology-elasticsearch-vm/</guid>
      <description><![CDATA[Setting up Elasticsearch 9.3 + Kibana on a Synology DS925+ by running Docker inside an Ubuntu VM to get seccomp support.<img src="https://feedpress.me/link/24086/17278655.gif" height="1" width="1"/>]]></description>
      <pubDate>Wed, 18 Feb 2026 00:00:00 GMT</pubDate>
    </item>
  </channel>
</rss>
