Back issues
Gergely Orosz on why Shopify is dropping React Native for native mobile six years after adopting it, with the reason named as AI agents: once one engineer with agents can build both platforms, the main benefit of cross-platform code shrinks. He covers why Shopify chose React Native in 2020 and quotes Shopify engineers on the change. Real reporting with named sources. The body opens with a book promotion, begins after the opening and stops at the cap, so the later sections are not here.
The New Stack on OpenAI's Decision API on GPT-6 Luna: models that return one of a fixed set of answers with a confidence score instead of chat text, which OpenAI says answer in 150 milliseconds against 1.6 seconds for Luna. The piece explains why prompted chat models give poor confidence scores for routing and classification. The latency figures are the vendor's own. The body begins after the opening.
Gergely Orosz on why Shopify is dropping React Native for native mobile six years after adopting it, with the reason named as AI agents: once one engineer with agents can build both platforms, the main benefit of cross-platform code shrinks. He covers why Shopify chose React Native in 2020 and quotes Shopify engineers on the change. Real reporting with named sources. The body opens with a book promotion, begins after the opening and stops at the cap, so the later sections are not here.
A data scientist revisits last year's predictions after two job changes: they have barely written SQL or Python by hand in six months, the role has moved from coder to reviewer, and repeated workflows are packaged as agent skills. The argument is that AI took the mechanical work and widened the job, with stakeholders self-serving more. First-person with concrete examples, no numbers beyond the author's own. The body stops at the cap and runs into related links.
The New Stack's write-up of GPT-6.1 Sol: same $2 and $10 per million token prices as Sol, near-Astra results at a fifth of Astra's cost, and an Ultrafast Codex mode up to eight times faster. Every figure is OpenAI's own benchmark chart, relayed. Tonight's pool also carries OpenAI's launch post, which is the primary. The site's owner invests in OpenAI and Anthropic.
Google Cloud launches GKE Agent Sandbox for agentic reinforcement learning, aimed at GPUs left idle while CPU sandboxes cold-start and multi-gigabyte images pull. The part worth a line is its claim that the team put Kubernetes itself on an agent-driven benchmark loop, using SWE-bench-style load to break its own clusters and feeding each bottleneck back. Vendor launch, vendor measurement.
A Faros executive on the three questions engineering heads keep asking: are we productive compared to what, what do we do once we know, and how do we change how teams work. The useful line is that the metrics that matter are the ones tied to your actual bottleneck, which differs by company. Thin, and a vendor selling the answer.
Simon Willison's live blog of OpenAI DevDay, first-person and timestamped. It covers Dots (always-on personal agents on GPT-6 Astra), ChatGPT Spaces, the Pro 500 plan, Ultrafast inference, the Decisions API he reads as a reply to TypeSafe's Jev, a cloud-only Codex and an Agents API with computer use. He notes Tejal Patwardhan saying OpenAI's models helped optimise their own computer-use harness for a 2x latency win, and that he switched from Codex Cloud to Claude Code for web that morning when Codex failed him. The body begins after the opening and stops at the cap, so the afternoon sessions are only partly here.
OpenAI launched GPT-6.1 Sol and says it nearly matches GPT-6 Astra on agentic coding and professional work at a fifth of Astra's price, with cached input at $0.10 per million tokens. All the benchmark numbers are OpenAI measuring itself. The useful line is at the bottom: an evaluation of whether an agent tells the user its search tool is broken instead of guessing, where Sol fails to say so 2.1% of the time and GPT-6 Luna 28.7%, on tasks chosen to provoke failure. The body begins after the opening.
Faros argues engineering metrics are heading for one headline number, time from customer insight to a shipped fix, with AI-adoption and PR-count metrics as transitional. Its sharper points: as AI absorbs a phase the bottleneck moves, sometimes within a quarter, from review to CI to release; speeding the inner loop can push problems into the outer loop; and the share of decisions escalated to a human is worth tracking because it shows where AI hits its confidence ceiling. A vendor that sells these metrics, citing its own research.
Honeycomb's product update: its Canvas agents can now read code, incidents, runbooks and tickets through connectors, with fleet-wide agent visibility and LLM cost tracking in early access and anomaly detection generally available. The framing, that AI-driven development brings more incidents that arrive faster and fail stranger, is the vendor's claim with no number behind it here. The sibling launch post on its AI Ecosystem product was blocked by our own guard tonight, so this row carries the only account of it.
Faros compares pull-request cycle times for 320 teams over a year, monorepo against polyrepo, and finds monorepo teams slower and more variable in practice. It then argues, without data, that the same properties may help coding agents reason about dependencies and make cross-cutting changes. The measurement is the vendor's own customer data; the agent claim is a hypothesis. The body begins after the opening.
Martin Fowler's short bliki entry on 'sensible default', the Thoughtworks term for a known-good practice you follow unless the context says otherwise and are ready to explain departing from, set against 'best practice', which implies no exceptions. He traces the term to Evan Bottcher and James Ross. Short, and a useful word for anyone writing team norms for AI-assisted work.
Simon Willison quotes Anthropic's Frontier Red Team on a binary-exploitation benchmark: on 100 randomly chosen tasks, the open-weight GLM-5.3 built full control-flow hijacks in 4% of trials against 6% for Claude Mythos Preview, and earlier models such as Opus 4.6 and GLM-5.2 managed none. The lab's own reading is that a threshold has been crossed and the capability is spreading beyond the frontier labs. It is a short quote from a longer post the pool does not carry, so the numbers are the lab measuring models including its own, relayed by Willison.
ByteByteGo's explainer on hallucination, built around a support bot that invents a 30-day refund window and a five-day payout. It separates factual, faithfulness and fabrication failures, explains why prediction produces confident inventions, and walks through RAG, tools and answer checking, ending on the point that citations only help if someone checks the citation supports the sentence it sits beside. Sound and clear, not new. It opens with a Sentry sponsor block and stops at the cap.
Nextgov on an SBA inspector general advisory: the agency piloted an AI fraud-detection tool on pandemic loans (a Palantir contract) without classing it as high-impact, which would have required minimum risk-management practices, and left it off its public AI inventory. The agency had already suspended 111,620 California borrowers in a fraud sweep. A watchdog finding that a consequential model ran without the safeguards its own rules require.
The Guardian on a freedom-of-information release: British Transport Police's six-month trial of live facial recognition in London stations scanned over half a million faces, cost £320,786 and nearly 100 officer hours, and produced one watchlist alert, which was a false match, and no arrests. The trial has since been extended to the Underground, and police report three confirmed alerts since. A clean example of a system judged by the number it was deployed to produce.
A Stack Overflow piece arguing a model version is not a release for an AI application: prompts, retrieval, tool contracts and serving settings change behaviour independently. It proposes a versioned release manifest, an evaluation gate that means something, and a rollback path that has actually been tested. The opening incident is labelled hypothetical by the author. Practical release discipline that applies directly to teams shipping agents.
OpenAI's launch post for Dots: always-on agents on GPT-6 Astra, each with its own cloud computer, plugins to over 4,000 apps, and memory of your preferences that grows with use. The examples are OpenAI's own: a Dot that starts investigating when a bug appears in Slack, one that turns a design into a working app. It is the house describing its own product, with no outside measure of how often a Dot's unsupervised work is right. The interesting question for a reader is the one the post skips: who reviews what the agent did while you were away.
The New Stack on what Dots mean for developers: agents that keep working when you step away, carry what they know across ChatGPT, Slack and Teams, and can watch customer feedback, scope fixes, build and test them, and return pull requests with videos of the change. All examples are OpenAI's. The article is clear on the claim and quiet on who reviews the pull requests and how often they are right. The site's owner invests in OpenAI.
Google's guide to graph workflows in its Agent Development Kit, worked through a refund example: start with one agent holding all the tools and the policy, then split into parallel lookups, a deterministic policy step, agent routers only where judgement is needed, a human review pause, and dynamic orchestration in Python. The useful point is when to take a decision away from the model and put it in code. A vendor tutorial, concrete and well built.
GitHub's developer policy update. The notable number is government takedown requests rising from 98 in all of 2025 to 708 in the first half of 2026, which GitHub says reflects counting every request, duplicates included, not a change in what it removes. The rest surveys the 2026 state legislative session and policy it is watching. A good small example of a figure that moved because the counting changed.
Ollama adds support for Jev-style decision models on a local endpoint: send state and named questions, get choices and scores in one request. It names three models and says Nimble 9B averaged 91 milliseconds per decision on an M5 Max, with accuracy compared on Bespoke Labs' benchmark suite. Vendor and partner numbers. The body begins after the opening.
The New Stack on OpenAI turning 'Sign in with ChatGPT' from an identity feature into a way to spend a Plus or Pro plan's allowance inside third-party tools, with per-partner limits. Sixteen launch partners are named, OpenClaw, OpenCode, Warp, Amp and Kilo Code among them. It matters to anyone paying twice for model access across tools. The body begins after the opening.
The New Stack on OpenAI cutting the $200 Pro plan's usage from 20 times Plus to 10 times from October 30, halving the GPT-6 Pro message allowance, and adding a $500 plan at 25 times Plus. Existing subscribers get a one-off credit grant that expires at year end. Tibo Sottiaux argues users will still get more done than a month ago; the quoted developer reaction is 'back to claude'. The site's owner invests in OpenAI and Anthropic.
Nextgov on a White House executive order telling agencies to call AI 'super intelligence' in official documents, with 60 days to propose a legal definition, and a frontier-safety accord signed the same day by major labs. The accord commits signatories to four layers of controls and audits, from monitoring capability and alignment during training onward. The body begins after the opening. The term is political; the accord's audit commitments are the part with consequences.
Today's paper: 25 stories, 5 on the front page, 6 desks staffed.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The Five, and it earned the spot — the wire scored it 7 of 10, I chose it.
Best enjoyed by anyone whose day holds more work than hours. The move: time it against the way you did it last week. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “AI Made Data Scientists Faster. Now It’s Expanding the Job.”; “OpenAI's new GPT-6.1 Sol undercuts its own Astra flagship”.
Pragmatic Engineer — 36 appearances in the Times since September 5, one of them today.
Pulled from Pragmatic Engineer. The original is worth your click — the byline earned it.
THE DIRECTING DESK today: 1 story.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
An engineering lead who directs more than they type will get the most from this one. The move: read it before the next review, not after. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE DIRECTING DESK — alone there today, which says something about the story.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
The New Stack has the full story; we hold the door open. Go read the original.
THE AMPLIFIER DESK today: 3 stories.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The Five, and it earned the spot — the wire scored it 7 of 10, I chose it.
Best enjoyed by anyone whose day holds more work than hours. The move: time it against the way you did it last week. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “AI Made Data Scientists Faster. Now It’s Expanding the Job.”; “OpenAI's new GPT-6.1 Sol undercuts its own Astra flagship”.
Pragmatic Engineer — 36 appearances in the Times since September 5, one of them today.
Pulled from Pragmatic Engineer. The original is worth your click — the byline earned it.
THE LOOP DESK today: 5 stories.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a builder running agents end to end. The move: let the agent read it before the next iteration. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “The top three problems engineering leaders keep running into”; “OpenAI DevDay 2026 live blog”, and 2 more.
Google Cloud AI Blog — 71 appearances in the Times since September 7, 2 of them today.
Google Cloud AI Blog has the full story; we hold the door open. Go read the original.
THE CONTEXT DESK today: 3 stories.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 5 of 10. Not the lead, and I'd still not leave it on the floor.
The one who feeds the agents' memory will get the most from this one. The move: file it where the agent looks first. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE CONTEXT DESK, beside 2 more today: “Monorepo vs Polyrepo: What the PR Benchmark Data Actually Shows”; “bliki: Sensible Default”.
Honeycomb Blog — 34 appearances in the Times since September 9, one of them today.
Honeycomb Blog has the full story; we hold the door open. Go read the original.
THE VERIFY DESK today: 5 stories.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 6 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a quality lead, or anyone who signs off. The move: sign off only after the evidence is on the desk. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “Why Do LLMs Lie?”; “SBA’s AI fraud detection pilot didn’t include needed safeguards, OIG says”, and 2 more.
Simon Willison — 144 appearances in the Times since August 22, 2 of them today.
Simon Willison has the full story; we hold the door open. Go read the original.
THE FLEET DESK today: 3 stories.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The Five, and it earned the spot — the wire scored it 6 of 10, I chose it.
Best enjoyed by a leader scaling past one agent. The move: pilot it on one team before the fleet. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 2 more today: “OpenAI just launched Dots. Here's why they matter for developers.”; “Graph Workflows in ADK: Everything You Need to Know | Google Cloud Blog”.
OpenAI News — 87 appearances in the Times since September 5, 2 of them today. Found via Hacker News Best.
Pulled from OpenAI News. The original is worth your click — the byline earned it.
THE WIRE today: 5 stories.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
Any reader building with agents will get the most from this one. The move: keep it for the next time the question comes up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
GitHub Blog AI — 40 appearances in the Times since September 5, one of them today.
GitHub Blog AI has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
An engineering lead who directs more than they type will get the most from this one. The move: read it before the next review, not after. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE DIRECTING DESK — alone there today, which says something about the story.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
The New Stack has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The Five, and it earned the spot — the wire scored it 7 of 10, I chose it.
Best enjoyed by anyone whose day holds more work than hours. The move: time it against the way you did it last week. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “AI Made Data Scientists Faster. Now It’s Expanding the Job.”; “OpenAI's new GPT-6.1 Sol undercuts its own Astra flagship”.
Pragmatic Engineer — 36 appearances in the Times since September 5, one of them today.
Pulled from Pragmatic Engineer. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The Five, and it earned the spot — the wire scored it 6 of 10, I chose it.
Best enjoyed by anyone whose day holds more work than hours. The move: time it against the way you did it last week. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “Why has Shopify dropped React Native?”; “OpenAI's new GPT-6.1 Sol undercuts its own Astra flagship”.
Towards Data Science — 101 appearances in the Times since September 5, one of them today.
Pulled from Towards Data Science. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
Anyone whose day holds more work than hours will get the most from this one. The move: pass it to whoever is out of hours. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “Why has Shopify dropped React Native?”; “AI Made Data Scientists Faster. Now It’s Expanding the Job.”.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
The New Stack has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a builder running agents end to end. The move: let the agent read it before the next iteration. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “The top three problems engineering leaders keep running into”; “OpenAI DevDay 2026 live blog”, and 2 more.
Google Cloud AI Blog — 71 appearances in the Times since September 7, 2 of them today.
Google Cloud AI Blog has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Inside pages at 4 of 10. The lead had a bigger day; this one had a point.
One for a builder running agents end to end. The move: put it in the loop and watch the second pass. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “Accelerate agentic RL with GKE Agent Sandbox | Google Cloud Blog”; “OpenAI DevDay 2026 live blog”, and 2 more.
Faros AI Blog — 47 appearances in the Times since September 5, 3 of them today.
Pulled from Faros AI Blog. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Today's Five, front page: the wire scored it 7 of 10; the stamp is mine.
This lands hardest with a builder running agents end to end. The move: run it end to end once before believing it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “Accelerate agentic RL with GKE Agent Sandbox | Google Cloud Blog”; “The top three problems engineering leaders keep running into”, and 2 more.
Simon Willison — 144 appearances in the Times since August 22, 2 of them today.
Our page is the summary; Simon Willison's is the story. Click through — they earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 6 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a builder running agents end to end. The move: let the agent read it before the next iteration. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “Accelerate agentic RL with GKE Agent Sandbox | Google Cloud Blog”; “The top three problems engineering leaders keep running into”, and 2 more.
OpenAI News — 87 appearances in the Times since September 5, 2 of them today. Found via Hacker News Best.
OpenAI News has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Today's Five, front page: the wire scored it 6 of 10; the stamp is mine.
Made for a builder running agents end to end, whether they know it yet or not. The move: run it end to end once before believing it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “Accelerate agentic RL with GKE Agent Sandbox | Google Cloud Blog”; “The top three problems engineering leaders keep running into”, and 2 more.
Faros AI Blog — 47 appearances in the Times since September 5, 3 of them today.
Our page is the summary; Faros AI Blog's is the story. Click through — they earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 5 of 10. Not the lead, and I'd still not leave it on the floor.
The one who feeds the agents' memory will get the most from this one. The move: file it where the agent looks first. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE CONTEXT DESK, beside 2 more today: “Monorepo vs Polyrepo: What the PR Benchmark Data Actually Shows”; “bliki: Sensible Default”.
Honeycomb Blog — 34 appearances in the Times since September 9, one of them today.
Honeycomb Blog has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
This lands hardest with the one who feeds the agents' memory. The move: feed it to the agents' memory before the next prompt. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE CONTEXT DESK, beside 2 more today: “Canvas Connectors & Fleet-wide AI Agent Visibility, & More”; “bliki: Sensible Default”.
Faros AI Blog — 47 appearances in the Times since September 5, 3 of them today.
Our page is the summary; Faros AI Blog's is the story. Click through — they earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The wire scored it 4 of 10 — inside pages, and worth the ink.
This lands hardest with the one who feeds the agents' memory. The move: feed it to the agents' memory before the next prompt. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE CONTEXT DESK, beside 2 more today: “Canvas Connectors & Fleet-wide AI Agent Visibility, & More”; “Monorepo vs Polyrepo: What the PR Benchmark Data Actually Shows”.
Martin Fowler — 26 appearances in the Times since August 24, one of them today.
Our page is the summary; Martin Fowler's is the story. Click through — they earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 6 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a quality lead, or anyone who signs off. The move: sign off only after the evidence is on the desk. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “Why Do LLMs Lie?”; “SBA’s AI fraud detection pilot didn’t include needed safeguards, OIG says”, and 2 more.
Simon Willison — 144 appearances in the Times since August 22, 2 of them today.
Simon Willison has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Inside pages at 5 of 10. The lead had a bigger day; this one had a point.
One for a quality lead, or anyone who signs off. The move: ask what proof would have caught it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “A quote from Anthropic Frontier Red Team”; “SBA’s AI fraud detection pilot didn’t include needed safeguards, OIG says”, and 2 more.
ByteByteGo — 38 appearances in the Times since August 22, one of them today.
Pulled from ByteByteGo. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
Made for a quality lead, or anyone who signs off, whether they know it yet or not. The move: write the test before the fix. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “A quote from Anthropic Frontier Red Team”; “Why Do LLMs Lie?”, and 2 more.
Nextgov AI — 48 appearances in the Times since September 5, 2 of them today.
Our page is the summary; Nextgov AI's is the story. Click through — they earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Inside pages at 4 of 10. The lead had a bigger day; this one had a point.
Best enjoyed by a quality lead, or anyone who signs off. The move: ask what proof would have caught it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “A quote from Anthropic Frontier Red Team”; “Why Do LLMs Lie?”, and 2 more.
theguardian.com — 3 appearances in the Times since September 30, one of them today. Found via Hacker News Best.
Pulled from theguardian.com. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 6 of 10. Not the lead, and I'd still not leave it on the floor.
A quality lead, or anyone who signs off will get the most from this one. The move: sign off only after the evidence is on the desk. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “A quote from Anthropic Frontier Red Team”; “Why Do LLMs Lie?”, and 2 more.
Stack Overflow Blog — 36 appearances in the Times since September 5, one of them today.
Stack Overflow Blog has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The Five, and it earned the spot — the wire scored it 6 of 10, I chose it.
Best enjoyed by a leader scaling past one agent. The move: pilot it on one team before the fleet. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 2 more today: “OpenAI just launched Dots. Here's why they matter for developers.”; “Graph Workflows in ADK: Everything You Need to Know | Google Cloud Blog”.
OpenAI News — 87 appearances in the Times since September 5, 2 of them today. Found via Hacker News Best.
Pulled from OpenAI News. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Inside pages at 5 of 10. The lead had a bigger day; this one had a point.
One for a leader scaling past one agent. The move: pilot it on one team before the fleet. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 2 more today: “Introducing dots”; “Graph Workflows in ADK: Everything You Need to Know | Google Cloud Blog”.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Pulled from The New Stack. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 5 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a leader scaling past one agent. The move: count what it costs before it scales. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 2 more today: “Introducing dots”; “OpenAI just launched Dots. Here's why they matter for developers.”.
Google Cloud AI Blog — 71 appearances in the Times since September 7, 2 of them today.
Google Cloud AI Blog has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
Any reader building with agents will get the most from this one. The move: keep it for the next time the question comes up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
GitHub Blog AI — 40 appearances in the Times since September 5, one of them today.
GitHub Blog AI has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The wire scored it 4 of 10 — inside pages, and worth the ink.
Made for any reader building with agents, whether they know it yet or not. The move: read it before the day fills up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
Ollama Blog — 14 appearances in the Times since September 9, one of them today.
Our page is the summary; Ollama Blog's is the story. Click through — they earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
Made for any reader building with agents, whether they know it yet or not. The move: read it before the day fills up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Our page is the summary; The New Stack's is the story. Click through — they earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
Made for any reader building with agents, whether they know it yet or not. The move: read it before the day fills up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Our page is the summary; The New Stack's is the story. Click through — they earned it.
The presses were quiet this morning — this is the September 30 edition, and it still holds up.
Inside pages at 5 of 10. The lead had a bigger day; this one had a point.
One for any reader building with agents. The move: share it with whoever will ask about it first. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
Nextgov AI — 48 appearances in the Times since September 5, 2 of them today.
Pulled from Nextgov AI. The original is worth your click — the byline earned it.
I'm missing a piece. Give me a minute.
Pragmatic Engineer: 36 in our pages since September 5. I read them all — emphasis on investigative.
I pulled this from Pragmatic Engineer. They've got the details.
I've got two questions already. I need better answers.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
Go read The New Stack. They earned the byline.
I'm missing a piece. Give me a minute.
Pragmatic Engineer: 36 in our pages since September 5. I read them all — emphasis on investigative.
I pulled this from Pragmatic Engineer. They've got the details.
Nothing in the files on this yet. I want to know why.
Google Cloud AI Blog: 71 in our pages since September 7. I read them all — emphasis on investigative.
This one's theirs. Read Google Cloud AI Blog's reporting.
The timing caught my eye. I'm checking that next.
Honeycomb Blog: 34 in our pages since September 9. I read them all — emphasis on investigative.
Go read Honeycomb Blog. They earned the byline.
Something changed. I want to know what.
Simon Willison: 144 in our pages since August 22. I read them all — emphasis on investigative.
This one's theirs. Read Simon Willison's reporting.
Not enough to call it yet. Back to the notebook.
OpenAI News: 87 in our pages since September 5. I read them all — emphasis on investigative.
I pulled this from OpenAI News. They've got the details.
Not enough to call it yet. Back to the notebook.
GitHub Blog AI: 40 in our pages since September 5. I read them all — emphasis on investigative.
Read GitHub Blog AI's original. They did the reporting.
I've got two questions already. I need better answers.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
Go read The New Stack. They earned the byline.
I'm missing a piece. Give me a minute.
Pragmatic Engineer: 36 in our pages since September 5. I read them all — emphasis on investigative.
I pulled this from Pragmatic Engineer. They've got the details.
There's a story here. We just don't have all of it yet.
Towards Data Science: 101 in our pages since September 5. I read them all — emphasis on investigative.
This came from Towards Data Science. Read their piece.
The headline is ahead of our files on this one.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
Read The New Stack's original. They did the reporting.
Nothing in the files on this yet. I want to know why.
Google Cloud AI Blog: 71 in our pages since September 7. I read them all — emphasis on investigative.
This one's theirs. Read Google Cloud AI Blog's reporting.
Our files stop short. I'll keep looking.
Faros AI Blog: 47 in our pages since September 5. I read them all — emphasis on investigative.
I can give you my notes. Faros AI Blog can give you the reporting.
I know what happened. I want to know what caused it.
Simon Willison: 144 in our pages since August 22. I read them all — emphasis on investigative.
Simon Willison has the full story. Start there.
From our research notes:
OpenAI News: 87 in our pages since September 5. I read them all — emphasis on investigative.
OpenAI News did the legwork on this one.
Same story as last edition. I'd go back to the original.
Faros AI Blog: 47 in our pages since September 5. I read them all — emphasis on investigative.
Don't stop with my notes. Faros AI Blog has the original.
The timing caught my eye. I'm checking that next.
Honeycomb Blog: 34 in our pages since September 9. I read them all — emphasis on investigative.
Go read Honeycomb Blog. They earned the byline.
No paper on the shelf has this one nailed down.
Faros AI Blog: 47 in our pages since September 5. I read them all — emphasis on investigative.
Faros AI Blog has the full story. Start there.
Something doesn't quite line up here.
Martin Fowler: 26 in our pages since August 24. I read them all — emphasis on investigative.
Martin Fowler has more than I can fit in this notebook.
Something changed. I want to know what.
Simon Willison: 144 in our pages since August 22. I read them all — emphasis on investigative.
This one's theirs. Read Simon Willison's reporting.
We don't have much on this one. I'm still looking.
ByteByteGo: 38 in our pages since August 22. I read them all — emphasis on investigative.
I can give you my notes. ByteByteGo can give you the reporting.
I checked the shelf. Nothing yet.
Nextgov AI: 48 in our pages since September 5. I read them all — emphasis on investigative.
Want the whole story? Nextgov AI has it.
I have the headline. I don't have the answer yet.
theguardian.com: 3 in our pages since September 30. I read them all — emphasis on investigative.
I pulled this from theguardian.com. They've got the details.
The timing caught my eye. I'm checking that next.
Stack Overflow Blog: 36 in our pages since September 5. I read them all — emphasis on investigative.
Go read Stack Overflow Blog. They earned the byline.
Not enough to call it yet. Back to the notebook.
OpenAI News: 87 in our pages since September 5. I read them all — emphasis on investigative.
I pulled this from OpenAI News. They've got the details.
There's a gap in the record here.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
I can give you my notes. The New Stack can give you the reporting.
I wrote this one up last time. Nothing useful to add today.
Google Cloud AI Blog: 71 in our pages since September 7. I read them all — emphasis on investigative.
This one's theirs. Read Google Cloud AI Blog's reporting.
Not enough to call it yet. Back to the notebook.
GitHub Blog AI: 40 in our pages since September 5. I read them all — emphasis on investigative.
Read GitHub Blog AI's original. They did the reporting.
The timing caught my eye. I'm checking that next.
Ollama Blog: 14 in our pages since September 9. I read them all — emphasis on investigative.
Want the whole story? Ollama Blog has it.
I have the headline. I don't have the answer yet.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
Don't stop with my notes. The New Stack has the original.
I checked the shelf. Nothing yet.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
Want the whole story? The New Stack has it.
From our research notes:
Nextgov AI: 48 in our pages since September 5. I read them all — emphasis on investigative.
I can give you my notes. Nextgov AI can give you the reporting.