<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
        <title><![CDATA[ mjsdotsocial - pckt ]]></title>
        <link><![CDATA[ https://mjsdotsocial.pckt.blog ]]></link>
        <description><![CDATA[  ]]></description>
        <language>en</language>
        <pubDate>Sun, 22 Nov 2026 18:43:20 +0000</pubDate>

                    <item>
                <title>Rented Competence</title>
                <link>https://mjsdotsocial.pckt.blog/rented-competence-radeuz7</link>
                <description><![CDATA[Why can&#039;t we just use this? I&#039;ve heard that question more in the last six months than in the previous ten years. In the era of GenAI, it&#039;s a fair question. The AI-built site or app or chatbot works. It was demoed to a client in a pitch and everyone saw it working, so it must be real. But here&#039;s the thing: it&#039;s the equivalent of a movie set. A movie set only has to look right for the camera. Walk around the back and you&#039;ll see plywood, paint, braces, and sandbags. A demo only has to work along it...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">rented-competence-radeuz7</guid>
                <pubDate>Wed, 11 Nov 2026 02:48:19 +0000</pubDate>
                            </item>
                    <item>
                <title>AI&#039;s effect on the software ecosystem</title>
                <link>https://mjsdotsocial.pckt.blog/ais-effect-on-the-software-ecosystem-q14r9a9</link>
                <description><![CDATA[If your day job isn&#039;t focused on software or application development, you&#039;re aware that AI has had a dramatic impact, but you might not realize the extent of its ripple effect on the ecosystem. It&#039;s not just code production that&#039;s transforming; it&#039;s the tools themselves, too. Tools, like the source code management application Git, were built for and around how humans work, but agents work differently than people do. Anthropic ran into this issue, and decided to build their own source code manage...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">ais-effect-on-the-software-ecosystem-q14r9a9</guid>
                <pubDate>Sun, 08 Nov 2026 19:57:03 +0000</pubDate>
                            </item>
                    <item>
                <title>You need a data plan</title>
                <link>https://mjsdotsocial.pckt.blog/you-need-a-data-plan-pdcvfky</link>
                <description><![CDATA[If you’re focusing your resources on building your agentic layer and your plan is to just connect it to your existing data as-is, you’re doing it wrong. Does your agent know where the data is and how to access it with the right permissions? Your data is sitting across multiple data platforms (Databricks, Snowflake, etc.), hyperscalers (Amazon Web Services (AWS), Google Cloud, Microsoft Azure, etc.), and enterprise systems ( Microsoft 365, Workday, Adobe Workfront, etc.). You can federate it, but...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">you-need-a-data-plan-pdcvfky</guid>
                <pubDate>Fri, 30 Oct 2026 15:25:22 +0000</pubDate>
                            </item>
                    <item>
                <title>AI Trends to Watch</title>
                <link>https://mjsdotsocial.pckt.blog/ai-trends-to-watch-tc86qge</link>
                <description><![CDATA[As your team builds your proprietary agentic runtime, two trends to watch: 1. Model macro-ization and micro-ization: Google DeepMind just released Genie 4, a world model capable of generating photorealistic, interactive environments with native audio and conversational characters for games and robotics. It&#039;s a shift toward using world models for creative direction, and while it&#039;s still very early, it&#039;s already encroaching on how agencies think about campaign creation. Liquid AI released LFM3, a ...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">ai-trends-to-watch-tc86qge</guid>
                <pubDate>Thu, 29 Oct 2026 15:38:02 +0000</pubDate>
                            </item>
                    <item>
                <title>AI voice innovation: Dessa</title>
                <link>https://mjsdotsocial.pckt.blog/ai-voice-innovation-dessa-hb2ct4e</link>
                <description><![CDATA[I&#039;d usually wait until a weekday to post, but I really think you should hear the latest in AI voice innovation, Dessa, from Block: https://block.xyz/dessa]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">ai-voice-innovation-dessa-hb2ct4e</guid>
                <pubDate>Sun, 25 Oct 2026 02:27:13 +0000</pubDate>
                            </item>
                    <item>
                <title>Claude Duet joins the autonomous agent space</title>
                <link>https://mjsdotsocial.pckt.blog/claude-duet-joins-the-autonomous-agent-space-d6ww23r</link>
                <description><![CDATA[Anthropic has launched Claude Duet, its standalone, autonomous agent for iOS. Duet joins the growing persistent agent space, proactively monitoring apps, prioritizing tasks, and integrating with commerce services like Stripe to execute purchases on behalf of its user. This move aligns Anthropic to the broader AI ecosystem trajectory, especially enterprise-targeted platforms like Deel with their AI Workforce, as just one small example. As these agents gain direct access and control across a user&#039;...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">claude-duet-joins-the-autonomous-agent-space-d6ww23r</guid>
                <pubDate>Sun, 25 Oct 2026 02:26:19 +0000</pubDate>
                            </item>
                    <item>
                <title>RSI coming to a harness near you…</title>
                <link>https://mjsdotsocial.pckt.blog/rsi-coming-to-a-harness-near-you-981cs7z</link>
                <description><![CDATA[Is your team still building a proprietary agentic harness or runtime? Does it include RSI? If not, add it to your ever-growing backlog. Google Research just open-sourced its &#039;Harness2&#039; recursive self-improving agent harness, enabling agents to encounter previously unseen tasks, propose, probe, refine, and improve itself, all while keeping model weights fixed. Check out the repo on GitHub: https://lnkd.in/gNRvVWMZ]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">rsi-coming-to-a-harness-near-you-981cs7z</guid>
                <pubDate>Sun, 25 Oct 2026 02:25:23 +0000</pubDate>
                            </item>
                    <item>
                <title>System One Extraction Models</title>
                <link>https://mjsdotsocial.pckt.blog/system-one-extraction-models-q1cz9a9</link>
                <description><![CDATA[Heads-up: Another new System One model, this time from Cloudflare. As announced on their blog, Clairic is a System One extraction model, built to extract values directly from unstructured inputs. Check it out: https://lnkd.in/gdCVfYTJ]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">system-one-extraction-models-q1cz9a9</guid>
                <pubDate>Sun, 25 Oct 2026 02:24:06 +0000</pubDate>
                            </item>
                    <item>
                <title>Perplexity on your Mac?</title>
                <link>https://mjsdotsocial.pckt.blog/perplexity-on-your-mac-57m5v73</link>
                <description><![CDATA[Perplexity Computer can now run fully local on your Mac. Work with your native Mac apps and local files, and run tasks without sending sensitive data to the cloud. The model and the agent both run on your Mac. Local tasks use PPLX Pro, our post-trained version of Qwen 3.8, and don&#039;t consume Computer credits. Available now to Pro, Max, and Enterprise subscribers on supported Apple silicon Macs. 36 GB of unified memory is recommended. Download the Perplexity Mac app and set up a local model to get...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">perplexity-on-your-mac-57m5v73</guid>
                <pubDate>Sun, 25 Oct 2026 02:23:23 +0000</pubDate>
                            </item>
                    <item>
                <title>AI Infrastructure Consolidation</title>
                <link>https://mjsdotsocial.pckt.blog/ai-infrastructure-consolidation-q1bt9a9</link>
                <description><![CDATA[More signs that the AI infrastructure sector&#039;s consolidation of independent neocloud providers are starting to crop up. One example: CoreWeave&#039;s bonds entered distressed territory last week, triggering a stock decline, and raising concerns about whether these neocloud buildouts represent sustainable investments.]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">ai-infrastructure-consolidation-q1bt9a9</guid>
                <pubDate>Sun, 25 Oct 2026 02:22:27 +0000</pubDate>
                            </item>
                    <item>
                <title>DHS - DeepSeek</title>
                <link>https://mjsdotsocial.pckt.blog/dhs-deepseek-4p0ya1t</link>
                <description><![CDATA[Heads-up: DHS is ordering the removal of DeepSeek products and third-party services from civilian agency information technology citing multiple threat vectors to systems. Read the order, here: https://lnkd.in/gvtH5TvG]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">dhs-deepseek-4p0ya1t</guid>
                <pubDate>Sun, 25 Oct 2026 02:21:54 +0000</pubDate>
                            </item>
                    <item>
                <title>From tokenmaxxing to self-replication</title>
                <link>https://mjsdotsocial.pckt.blog/from-tokenmaxxing-to-self-replication-qct39a9</link>
                <description><![CDATA[We had tokenmaxxing, then the tokenpocalypse, and then the token famine. This drove price wars, interest and competition in open source/weight models, new model types (e.g, decision, CDM, etc.), and the pursuit of locally-runnable models. As innovation, including LLM gateways like OpenRouter, continues to drive capability, capacity, and cost efficiency for the hosted model ecosystem, the same is happening for local models, and not just for running open models (e.g., Qwen Coder, etc.), locally on...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">from-tokenmaxxing-to-self-replication-qct39a9</guid>
                <pubDate>Mon, 19 Oct 2026 21:14:55 +0000</pubDate>
                            </item>
                    <item>
                <title>Pay attention to the AI company shakeout</title>
                <link>https://mjsdotsocial.pckt.blog/pay-attention-to-the-ai-company-shakeout-4jeka1t</link>
                <description><![CDATA[Pay attention: as consumer-play AI companies like Character.ai (who filed for bankruptcy after ~5 years) are winnowed out, watch for AI startups to pivot toward the enterprise, or abandon their own stacks in favor of companies chasing the enterprise like Anthropic or OpenAI, who have the focus, innovation, and scale, to succeed. It&#039;s a familiar story.]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">pay-attention-to-the-ai-company-shakeout-4jeka1t</guid>
                <pubDate>Mon, 19 Oct 2026 21:13:42 +0000</pubDate>
                            </item>
                    <item>
                <title>Another new model architecture just dropped</title>
                <link>https://mjsdotsocial.pckt.blog/another-new-model-architecture-just-dropped-8g1qnwy</link>
                <description><![CDATA[Another new model architecture just dropped. According to ScaleDown, Constrained Decoding Models (CDM) are task-specific, purpose-built language models, each trained to do one job: compression, summarization, extraction, and classification. Check out the post from founder Neal Patel: https://lnkd.in/gbDE-yWn]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">another-new-model-architecture-just-dropped-8g1qnwy</guid>
                <pubDate>Mon, 19 Oct 2026 21:13:06 +0000</pubDate>
                            </item>
                    <item>
                <title>You better move quickly</title>
                <link>https://mjsdotsocial.pckt.blog/you-better-move-quickly-h2f4t4e</link>
                <description><![CDATA[If you want the agentic runtime your team is developing to stay at parity with those offered by AI companies, you&#039;ll need to move quickly. Always-on, user-dedicated autonomous agents (Spark, Muse, Dots, Grok Bot, etc.), dynamic work analysis with model autorouting (monomodal, multimodal, omnimodal; decisioning, etc.), cost management (frontier, flagship, mid-tier, lightweight, open weight/source, etc.), data (ontologies, semantics, taxonomies, etc.), DBs, (relational, vector, graph, KV, multimod...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">you-better-move-quickly-h2f4t4e</guid>
                <pubDate>Mon, 19 Oct 2026 21:12:13 +0000</pubDate>
                            </item>
                    <item>
                <title>Your agentic runtime is already antiquated</title>
                <link>https://mjsdotsocial.pckt.blog/your-agentic-runtime-is-already-antiquated-fdpyhyp</link>
                <description><![CDATA[The enterprise agentic runtime your team is building was likely built on a model of how people work in the enterprise: handling individual tasks. That evolved to serial chained tasks, and then, maybe, to semi-autonomous workflows where the agent decides what to do based on the input and training of your SMEs. The AI companies have already evolved past task-based workflows. Whats next? Always-on agents, an enterprise evolution of OpenClaw. Backed by a persistent user-dedicated secure VM or worksp...]]></description>
                <author>mjsdotsocial</author>
                <guid isPermaLink="false">your-agentic-runtime-is-already-antiquated-fdpyhyp</guid>
                <pubDate>Mon, 19 Oct 2026 21:10:31 +0000</pubDate>
                            </item>
            </channel>
</rss>