Short notes, quick updates, and things I find worth sharing. For the full graph of my thinking, browse my public notes. Subscribe via RSS.
OpenHarness (the app itself is just called "Harness") is an open-source command center for coding agents, built by Autonomous (autonomous.ai). Their tagline: "The ultimate harness for coding agents and beyond. All your agents. All your machines. One command center."
GPT-6.1 Sol is the middle model of OpenAI's GPT-6 family, between GPT-6 Luna and GPT-6 Astra. OpenAI released it on 29 September 2026, at DevDay, exactly one week after GPT-6 Sol. It replaces GPT-6 Sol as "the newer Sol model" in the API docs, so GPT-6 Sol had the shortest run as OpenAI's current So
Before you type anything, a Claude Code session already loads the system prompt, tools, MCP servers, agents, memory files and skills. All of it takes space in the same Context Window, and all of it is sent with every request.
I just shipped Obsidian Starter Kit 5.0.1. After v5.0.0 went out on September 25, some of you told me that the daily note template wasn't always applied. You were right, and the cause was a change in Templater.
RLCD is how TypeSafe AI says it trains Jev. The name stands for Reinforcement Learning for Calibrated Decisions, and the pitch fits in one line from their docs: RLCD "trains TypeSafe to return decisions and calibrated probabilities instead of generated text."
I just released a new plugin: Agentic Resource Discovery Server plugin for Obsidian. It turns your vault into a local registry for your AI skills and agents. Your AI assistant searches that registry when it needs something, and fetches only that.
A decision model (DM) is an AI model whose output is a decision, not text. You give it an input and a bounded set of possible answers, and it gives you back numbers: which option, how likely yes is, what score. It's the category Jev belongs to, and I think it deserves its own name next to Large Lang
A System One model is TypeSafe AI's name for a general-purpose decision model). It takes context and a set of typed questions and returns probabilities: which option, yes or no, what score. It doesn't generate text. The name comes from Daniel Kahneman's two systems in Thinking, Fast and Slow): Syste
fast-jev-compaction is an MIT-licensed plugin for Claude Code by Tamara Tran that replaces summary-based compaction with Jev decisions. The question behind it is a good one: why is compaction still a summarization prompt in 2026?
Projects is a Claude Code feature launched in beta on 2026-09-17, on desktop and web (cloud sessions), for select Pro and Max users. Anthropic says it's coming to all Claude users soon. By early October 2026 the docs describe it as a public beta on Pro and Max.
Less than two months after Claude Opus 5, Anthropic just released Claude Opus 5.5. It's the first model of the new Claude 5.5 family, and Sonnet 5.5 and Haiku 5.5 are coming "in the coming weeks".
When you run several Claude Code sessions in parallel, YOU are the orchestrator: you open five sessions, try to remember which one does what, and copy context from one to the other by hand.
Some of the LLM calls in an AI agent or automation are really decisions: Is this email urgent? Which tool should I call? Is this paragraph off-topic? Is the agent stuck?
I'm using Git worktrees more and more, mostly because every parallel coding agent needs its own working directory. And sooner or later, you hit the question from this Stack Overflow thread: I changed something in worktree A, how do I get it into worktree B? Is there a `git worktree pull`?
Here's a little-known tip for Obsidian. Images can be placed side by side in obsidian, if they are on the same line and their size allows them to. If you have a first image with a width of 200px and a second one a width of 300px, you can use `` to display those side by side, if your note has at leas
Software for one is software built by and for a single person (or a household, or a tiny team), designed around their exact needs and workflow, with no intention of scaling, selling, or pleasing anyone else. Its value comes from fitting one context perfectly, not from reaching many users.
GitHub Radar is a free, open source (MIT) dashboard that keeps every open GitHub issue and pull request across the users, organizations and repositories you care about on one page. I built it to track my own projects without hopping between dozens of repositories, but it is generic: it starts empty
fx is a coding agent CLI from Vercel Labs, written in Zig. The pitch fits in one line: "Tiny, open, embeddable, native coding agent." The whole thing is a ~6 MB binary, Apache-2.0 licensed, and it runs on macOS and Linux. It's still experimental (v0.0.x at the time of writing), and Vercel says it pl
CSS is getting a selector that targets every class sharing a prefix, such as `.btn-`.
I just released a new plugin: Knowii Community plugin for Obsidian. It brings the Knowii Community right into Obsidian. You can read, post, reply and get notified without ever leaving your vault.
Isham Faizal published a small experiment on September 7, 2026: an LLM attention visualizer. You hover over (or tap) any token a model generated, and the earlier tokens that fed into it light up.
Discord Sync CLI (`discord-sync`) is a CLI/TUI I built to export and synchronize Discord servers, channels, and threads to JSON, HTML & Markdown. Years of community knowledge live in Discord, and none of it is really yours: no offline access, no proper export, and attachment links that silently expi
Three months after the launch (and the ban, and the un-ban) of Claude Fable 5, Anthropic just shipped Claude Fable 5.1 and Mythos 5.1.
The Time Machine plugin for Obsidian just got its biggest update so far: 2.0.0.
Five releases of the Kanban Action Planner landed today, from 1.18.0 to 1.19.2. They add column triage, focus mode, time tracking, new filter aliases, and a performance fix for large boards.
The Dataview Serializer plugin for Obsidian just reached 3.0.0.
The way we learn and perceive software development is changing a lot due to the rapid evolution of AI capabilities.
You Think Faster Than You Type Your Keyboard Is Costing You Hours Every Week
Aliases give a note alternative names. They are defined in the note's frontmatter (see Obsidian Properties) and picked up everywhere Obsidian resolves names: wikilink autocomplete, the Obsidian Quick Switcher, and unlinked mentions.
Embeds display the content of another file inside the current note. The syntax is a wikilink prefixed with `!`. This is Obsidian's implementation of Transclusion: content lives in one place and gets reused by reference, never copied.
Obsidian search (backed by the Search core plugin for Obsidian) is full-text search across the whole vault, with operators that most users never discover. Learning a handful of them is the difference between scrolling through results and getting exactly the notes you meant.
The way I use AI has gone through many iterations already.
Model routing is the practice of directing different tasks to different AI models based on the task's requirements. Instead of using one model for everything, a routing layer selects the most appropriate model considering capability, cost, speed, and context needs.
Microsoft eXecution Container (MXC) is an open-source (MIT License) sandboxed code execution system from Microsoft for running untrusted code (model outputs, plugins, tools) safely on Windows, Linux, and macOS. Instead of shipping one isolation technology, it puts multiple containment backends, from
Zellij is a modern terminal multiplexer written in Rust. It positions itself as a "terminal workspace" that doesn't sacrifice simplicity for power; discoverable UI with on-screen keybinding hints makes it approachable without limiting what power users can do.
GHE.com is the version of GitHub Enterprise Cloud with data residency: a managed cloud GitHub where a company chooses the region its code and data are stored in. Each enterprise gets a dedicated subdomain (e.g. `octocorp.ghe.com`) with its own namespace, fully isolated from the open-source community
Awesome GitHub Copilot (awesome-copilot.github.com) is GitHub's official, community-driven catalog of resources that extend GitHub Copilot. Maintained by GitHub with community contributions, backed by the `github/awesome-copilot` repository. It's the de facto marketplace for Copilot customizations a
Canvases are durable, shared visual surfaces in the GitHub Copilot App where you and an AI agent operate on the same plan. Instead of workflow state living buried in chat history, a canvas makes it explicit and persistent: phases, decision points, validation gates, and drafts update live as the agen
GitHub Copilot MCP Registry is the ability, in public preview, to point a GitHub organization or enterprise at a private MCP) registry so that GitHub Copilot clients only discover, and optionally only allow, the MCP servers your company approves. Admins configure it under AI Controls → MCP in the or
GitHub Copilot Sandboxing is GitHub's answer to the "agent with full shell access" problem: two complementary isolation modes for running GitHub Copilot CLI agent sessions, both in public preview. Local sandboxing wraps the commands an agent runs on your machine in an OS-level sandbox powered by Mic
Programs that let you run multiple terminal sessions inside a single terminal window or remote connection. Key capabilities: split panes, multiple windows/tabs, session persistence (detach/reattach), and remote session survival (SSH disconnects don't kill your processes).
Callouts are styled block elements in Obsidian that highlight important information using a special blockquote syntax. They replace the older Admonition plugin for Obsidian approach with a native, built-in feature.
Obsidian URI is the custom `obsidian://` protocol that Obsidian registers with the operating system. It lets any external application (browser, launcher, task manager, script) open vaults, jump to notes, run searches, and create content. No plugin needed; this is the official way to control Obsidian
A new open specification landed: Agent Plugins 1.0, with an initial technical steering committee made of Amazon, Cursor, Microsoft, OpenAI and Vercel. It defines a portable package format for extending agents, so an extension is authored once instead of being repackaged per client.
Remotely Save, synchronizes an Obsidian vault with cloud storage you already have: Dropbox, OneDrive, Google Drive, any WebDAV server, or S3-compatible object storage (AWS S3, Cloudflare R2, MinIO, Backblaze B2, and similar). It is a lightweight alternative to Obsidian Sync for people who prefer to
OpenAI reduced the Codex model's context size from 372k to 272k tokens. There was no announcement; the change appeared in a pull request in the open-source Codex CLI repo, and that pull request reached the Hacker News front page.
OpenRouter added speech-to-text to its API. A new endpoint (`POST /api/v1/audio/transcriptions`) accepts base64-encoded audio and returns the transcript as JSON, using the same API key as chat completions, so you don't need a separate STT provider or server.
The Notebook Navigator plugin for Obsidian just hit a milestone: it's now the #1 most popular plugin in the Obsidian community directory, and should pass a million downloads within weeks.
Give deterministic code everything it can handle (routing, scheduling, data transformation, file writes), and reserve the Large Language Models (LLMs)) for the parts that need judgment (interpreting ambiguous, unstructured input).
Luca Rossi makes a point about AI coding that matches my thinking: most AI coding workflows optimize the INNER loop, the rapid cycle of describing a feature and generating code. But that's a fraction of software delivery. The OUTER loop (deciding what to build, shaping the solution, releasing, learn
A clear roadmap to building your own knowledge system. Subscribe and get it straight to your inbox.
6,000+ readers. No spam. Unsubscribe anytime.
Every week, I share practical systems, tools, and frameworks to help you organize your knowledge, leverage AI, and create consistently. No fluff. No spam.
6,000+ readers already subscribed ✨
Every Knowii tier includes the private community, live events, expert access, monthly challenges, a 3.8K+ concept knowledge base and my ~10K published notes. Pick the depth that fits you:
€14.99/quarter
€59.99/quarter
€119.99/quarter
