Brent maybe try a closer camera angle in a nature background like

Nemesis system was actually AI. Checkmate Mister Zitron 🥲

Didn’t expect my F1 & AI interests to collide but 2026 battery regulations are forcing it! Here’s a fantastic explanation of narrow AI used for engine management by @eddstrawF1 that I just had to call it out as a correct way to articulate the topic: youtu.be/j0_Rb84XKgA?si=9yXR…

Recursive Self-Improvement (RSI) for software harnesses is having a moment this week with @PrimeIntellect’s self-improving Prime Agent and @sawyerhood’s great article on bb, and IDE that builds itself. Here are several more of the top articles I’ve been reading to learn about RSI for harnesses in particular: * lilianweng.github.io/posts/2… - Harness Eng for Self-Improvement is just a great in-depth article about harness design patterns, not tied to any specific implementation * metr.org/blog/2025-02-14-mea… METR’s measuring automated kernel engineering from early 2025 contains a lot of detail based on 4o-level models and shows how difficult it is to measure realistic tasks which are likely necessary for RSI’s feedback loops * normaltech.ai/p/ai-agents-ca… - not about harnesses specifically but a summary on a recent paper that covers what I’ve recently realized where agents are deciding based on data, and when the data isn’t available it’s a path not taken. Perhaps a “research harness” could course correct an AI model at the right time? Links to the articles I mentioned in the intro sentence: Prime Intellect’s article on Prime Agent: primeintellect.ai/blog/prime… Sawyer’s article on bb: xitter.cf/sawyerhood/status/2085… Image is from the Harness Engineering for Self-Improvement Post.

Loop engineering developer trends are now available in Plushcap. This trend tracks product features and content by dev tools companies for designing self-sustaining feedback loops around AI coding agents so they can re-prompt themselves, evaluate their own output, and iterate toward a goal without requiring a human to intervene at each step. "Ralph loops" and "Ralph Wiggum loops" are also included in this one, but not graph engineering, as that seems to be a closely related but still separate architectural branch for making AI coding tools more autonomous. Insights on loop engineering: plushcap.com/blog/loop-engin… Direct trend data link: plushcap.com/trends/monthly/… All of the data and insights are available as inputs to Claude/ChatGPT/etc via the Plushcap MCP server and API: mcp.plushcap.com

A few resources on local model coding agents (especially @googlegemma 4) that I'm really enjoying learning from recently: * simonpcouch.com/blog/2026-04… - 14-24B dense local models scored 0/10 on his refactoring eval but Gemma 4 26B-A4B jumped to 9/10. 26B parameters but MoE 4B active per token, so it's very fast. Matches my own experience on my 4090 RTX! * magazine.sebastianraschka.co… A great guide for wiring up local models to coding harnesses such as Claude Code, and goes into differences in models. For example, gemma4:e2b fails 0/5 on tool-reasoning tasks while Qwen 35B-A3B class solves 4-5/5 * patloeber.com/gemma-4-pi-age… - awesome walkthrough by @patlober that shows how to configure @pidotdev (which is also my local model coding harness of choice) with Gemma 4 * interconnects.ai/p/gemma-4-a… - argues that 30B parameter count is the number that matters (I disagree, 26B w/ MoE seems to be the floor for me) while 7B is for tinkering but overall a great read There seems to be a parameter floor in the ~26B model range below which local models just don't feel like as much of a breakthrough, and above that the best models like Gemma 4 and Qwen 3.x remind of of that "aha!" moment I had with Opus 4.5 early last year.

damn that’s an impressive subscribers trajectory: plushcap.com/companies/opena…

What an incredibly well-written, thoughtful, and informative article on data tools from a developer perspective, "Guide to data tools landscape for devs": sinja.io/blog/data-landscape… Reminds me of some of my other favorite technical posts in areas like storing & accessing data, how web browsers work, * queue.acm.org/detail.cfm?id=… for a sweeping view on progress in how storage is accessed and its behavior * howbrowserswork.com/ an interactive guide to how HTTP, DNS, SSL, TLS and all of the most important protocols work in the browser (TLS 1.3 is similarly great tls13.xargs.org/) * sinja.io/blog/bot-or-not - another great post from the same author but this one on bots & detecting them Amazing work @OlegWock, you deserve waaaay more followers and readers for your writings

Several articles & resources I’ve been reading lately to learn about how LLM architectures have evolved: * finbarr.ca/five-years-of-gpt… 5 years of GPT-class model progress, only goes to 2023 but that puts more focus on early models where details are precisely known * magazine.sebastianraschka.co… great companion post to the first one as its diagrams and code examples are absurdly helpful. @rasbt has so many articles that could be on this list but I picked this one because of the visuals in particular * gregorygundersen.com/blog/20… denser than other posts. my attention can glaze over some of the deeper math formulas but as a synthesis of key LLM concepts have evolved this is excellent * cameronrwolfe.substack.com/p… both accessible in the basics "what is a token?" and detailed, this one goes into other notable models from early LLM days (post image is from this article)

A few resources on harness engineering I'm really enjoying learning from recently: * walkinglabs.github.io/learn-… - excellent course that covers how to constrain agent behavior in a productive way, what the challenges are using LLMs especially maintaining context, and how to make the harness more agentic and work to the proper conclusion of a task * twotimespi.dev/ - build a simplified agent in Python (Tau). start with "uv tool install tau-ai" * humanlayer.dev/blog/skill-is… - lessons and learnings from failing and succeeding to build harnesses for coding agents over 12+ months * langchain.com/blog/the-anato… - higher-level overview of the many components that need to come together properly in a harness to make an agent work

New developer trend I'm tracking: harness engineering. It's still early and there's a ton of room for the trend to grow. Prediction: "It would not be surprising to see 10x features, posts and mentions for this topic by this time in 2027. It's simply impossible for AI agents to operate at scale without further learnings in the software harness layer, especially as agents interact with other agents and not just existing deterministic APIs." Report: plushcap.com/trends/reports/… Underlying trend data: plushcap.com/trends/weekly/h…

Build Your First MCP Server: Code Tips in Claude and Slack - fantastic Python tutorial by @bbelderbos on building a quick MCP server and testing it in @ClaudeDevs! belderbos.dev/blog/build-mcp…

Check out @nhost, an awesome backend-as-a-service company with GraphQL, PostgreSQL and Hasura at its core. Just added them to Plushcap based on a user request: plushcap.com/companies/nhost There's also a detailed product and content analysis report for them: plushcap.com/trends/reports/…

Re-downloaded from HF 30 mins ago, it’s working for me on Macbook M3 Max 128gb. Tried both w/MTP and w/o MTP while using Pi on some projects. Didn’t see any speed up from MTP (first shot is new q2 without MTP, second is with MTP), but no issues. Amazing project, keep it going!

Just added @OpenObserve to Plushcap (link: plushcap.com/openobserve). The company was founded in 2022 with a $10m Series A raised last month. They ramped technical content hard before that with 69 posts published in 2026 (256 blog posts all time). That output velocity is a typical pattern for a DevTools PLG company to make the seed to Series A jump. Their content strategy involves high output SEO/GEO and integration tutorials. They've published a 9-part "Datadog vs OpenObserve" series covering logs, metrics, traces, dashboards, alerts, RUM, pipelines, IAM, and cost. On top of that, dedicated "Top 10 alternatives" posts targeting Datadog, Grafana, Splunk, New Relic, and more. They're publishing integration tutorials across AWS, Azure, GCP, plus databases like PostgreSQL, MongoDB, and ClickHouse. More recently they pivoted into AI/LLM observability content including how to monitor OpenAI API costs, trace LangChain and LlamaIndex apps, observe MCP servers, and monitor production AI agents. YouTube is still early (960 subscribers, 91K views across 52 videos) but output matches the pace of their written output. Full data for blog output, YouTube metrics, competitive positioning, and website breakdown: lnkd.in/dVJUfZw5

Congrats to @exedev on their Series A funding blog.exe.dev/series-a Cloud infra is long overdue for a rethink and rebuild. There are many good reasons and ideas captured in @davidcrawshaw ’s manifesto: crawshaw.io/blog/building-a-…

"Can you reimagine plushcap.com as a 16-bit japanese RPG?" .... new GPT-image-2 is legit. Might need to go all in on this design for next version.