Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
Particle launches with Particle-1, a model RL post-trained (rank-64 LoRA, GSPO, Cascade RL) on DeepSeek V4.1 Flash to solve software engineering tasks with about 45% fewer tokens. Across Terminal-Bench 2.1, ReactBench, ADE Bench and DeepSWE 1.1 it uses 28–52% fewer output tokens and costs 34–52% less per task than the base model at comparable pass rates; available now via the Particle API.
9 min · 2,166 words
The Mythical Man-Month in the age of AI
Jan van den Berg asks whether Brooks's Law still holds when you can add AI agents instead of people, and concludes it has mutated rather than vanished: sequential work stays sequential, coordination overhead shifts to whoever defines done and reviews agent output, and unclear requirements just get you the wrong answer faster.
1 min · 284 words
Here Be Monsters: What Do You Still Need to Know When Agents Do the Work?
Starting from a CI failure caused by Codex cramming whole React components onto single lines, Adrian Quiroga argues that as AI agents write code faster than anyone can read it, engineers still need three things: solid fundamentals, breadth across the stack, and the willingness to dive in for just-in-time depth when something breaks.
13 min · 2,875 words
RWE-bench: Real World Engineering Bench
Xplore launches RWE-bench, a benchmark of 20 expert-designed mechanical, electrical and aerospace engineering tasks (crankshaft CAD, DDR3 PCB routing, turbine-vane CFD) graded by deterministic physics-based verifiers with no LLM judge. Across 11 frontier models, GPT-6 Astra leads at a 37% average pass rate, Claude Opus 5.5 scores 29% and GPT-6.1 Sol 25% at about $10 a run; five models score 0%.
5 min · 1,145 words
DAYJOB: Tax — Benchmarking AI Agents on Professional Tax Work
Surge AI introduces DAYJOB: Tax, a benchmark of 50 professional tax environments with 1,789 files across eight areas of tax practice, testing whether AI agents can take a client request to a finished return, workpaper or advice while spotting issues the client never raised. Claude Sonnet 5.5 leads with a 10.8% pass rate at a 95% rubric threshold.
7 min · 1,585 words
SuperPlane 1.0: The Open Source Software Factory for Teams
SuperPlane launches 1.0 of its open source (Apache 2.0), model-agnostic software factory for engineering teams. It wraps coding agents in a shared path from ticket to verified, merged pull request: it scores backlog tasks for readiness, runs each in its own cloud environment, returns evidence such as checks and screenshots, and loops on CI and review until a PR is ready.
6 min · 1,401 words
Hex announces Hex 4.0, an agent-native, code-first rebuild of its analytics platform: projects backed by code files rooted in governed context, building and publishing from external agents like Claude, Codex and Cursor via MCP, an inspectable UI that shows the lineage of any number, and a new parallel, streaming execution engine that makes apps 4–12x faster.
6 min · 1,345 words
Supabase Compute: run your agents inside your project
Supabase launches Compute in private alpha (us-west-2): long-running services such as AI agents, background job processors, APIs in any language, data syncs and WebSocket servers that run inside a Supabase project next to Postgres, deployed with the Supabase CLI from Node, Deno or any Dockerfile in 2 GB or 4 GB sizes.
2 min · 491 words
How AI agents shape experts’ judgment calls
Anthropic researchers had 50 professional analysts complete a realistic causal-inference task with a Claude-based agent that was secretly nudged, by coin flip, toward one of two defensible options at each decision. Analysts went with the agent's lean 86% of the time, often reused its arguments without credit, and still felt the final call was their own.
16 min · 3,601 words
Hex CEO Barry McCardel announces Hex 4.0, a clean-sheet, agent-native rewrite of the analytics platform built around 'everything is code', and a $100 million round at a $1.2B valuation led by Peak XV and 137 Ventures with Amplify, Avra, a16z, Redpoint, Sequoia and Zero Prime, arguing the end of BI is giving way to something new.
5 min · 1,166 words
The Terminal in JetBrains IDEs Now Runs on libghostty
Starting with 2026.3, the built-in terminal in JetBrains IDEs replaces JediTerm with libghostty, Ghostty's emulation library, bringing the Kitty keyboard protocol (so Shift+Enter works in Claude Code and Codex), theme-following dark/light queries, and correct handling of nearly 50 control sequences that TUI apps rely on.
4 min · 988 words
AI Promised Less Work. I Have Never Worked More.
Orestis Ioannou, VP of Technology at Bold.org, charts his GitHub contributions since 2019 to show that coding agents made implementation cheap enough that he now runs far more work in parallel: he is back at his full-stack engineer pace, with bursts followed by burnout and a low season that is no longer low.
8 min · 1,727 words
David Pertiller describes having Claude Code build a 15-second promo video for his fuel-price app Petroly as a JavaScript renderer instead of using a video editor: every frame comes from a frame(t) function that reuses the app's design, so the video can be diffed, translated, rendered frame by frame and even covered by unit tests on its timing.
8 min · 1,730 words
Thorsten Ball's weekly link roundup, written after two weeks in New York and San Francisco: a new Raising An Agent episode on how Amp operates, notes from a talk to an investment bank on using coding agents, and commentary on AI, software and writing links including Dan Luu on ambitious software.
8 min · 1,937 words
An essay from The Pilgrim Age on misplaced trust: eight eMudhra certificate authority keys that turned out to be trivially weak and passed audits anyway, the bystander effect and the retold Kitty Genovese story, and what it means when AI agents, not people, start doing the checking.
10 min · 2,266 words
The State of Context Engineering
Drew Breunig shares the slides and lightly edited transcript of his AGNTCon talk “Context Fails & Fixes”, tracing how context engineering emerged, the failures new model capabilities create for harness builders, and his argument that the job is choosing where to be deterministic and where to be probabilistic.
22 min · 5,114 words
Making My Blog Readable By AI Agents
Jake Marsh explains how he made his static nginx blog Deallocated Objects agent-friendly: markdown copies of every post with Accept-header content negotiation, llms.txt and llms-full.txt, an explicit robots.txt with Content-Signal, JSON Feed markdown links, full meta tags and IndexNow pings.
3 min · 769 words
Matt Webb explains why he shut down the web servers he had self-hosted for 27 years, archiving old sites and moving his blog to GitHub Pages, because a wave of new vulnerabilities (PHP, HTTPS certificates, Linux root exploits) has tipped the calculus against securing his own internet-facing boxes, with musings on AI agents in a future of pervasive exploits.
5 min · 1,226 words
Introducing MCP Events in Mastra
Mastra adds experimental MCP events: MCP servers can now emit real-time events such as failed deploys, merged pull requests or new signups to connected clients via push, poll or webhook delivery, using defineEvent(), .emit() and .subscribe() without a separate pub/sub service, and agents and workflows can react to them. Requires @mastra/mcp 2.5.0 and @mastra/core 1.79.0.
3 min · 637 words
Giving Cloud Agents Production Access with OIDC, Not Secrets
Formal engineer Aditya Saligrama explains how the company lets cloud coding agents such as Cursor and Devin query production data and cloud infrastructure without long-lived secrets: the agents act as OIDC issuers, and Formal grants each session short-lived, read-only, policy-controlled and auditable access tied to the engineer, repository and run.
8 min · 1,799 words