What's moving through AI, ranked by lead time.
Radar is a running index of what's moving through the AI ecosystem, ranked by where each story sits on the lead-time gradient: papers become repos become community threads become analysis. Every ingest run of the pipeline takes a fresh snapshot — several times a day, on no fixed clock — and the stamp below is the exact time of the one you're reading. The earlier a story is caught, the more head start you have before the crowd. Every row carries its receipts — which source saw it first, how many independent sources have it, how long ago — so the ranking explains itself. It's the public face of the same pipeline that feeds Vlad's Newsletter.
frozen 19 Nov 2026, 22:14 UTC · 67/78 feeds live · 16,356 tracked since 12 Jun
- 1 Rebuilding my local voice assistant pipeline blog.platypush.tech · 2h
- 2 PauseAI Munich Rally | Stop the loss of control over AI lesswrong.com · 18m
- 3 We need commitments for 25% compute for safety by 205 ECI points lesswrong.com · 30m
- 4 Show HN: Billet – 6MB (Agentic) Task Management System mktwsk.xyz · 54m
- 5 Show HN: A SEO Tool to get found by Google and get cited by AI mozolite.com · 1h
509 stories entered the watched feeds in the last 24 hours — 265 in papers, 7 in repos, 203 in community, 34 in analysis. The scored board skews to Community (40 of 40). 1 story is heating across independent sources, and the median lead across the ledger's 13 observed fires is 6d — measured, not estimated. Freshest receipt, 12d ago: caught in GitHub Trending Python 119d before Show HN carried it. The oldest survivor has held the board for 5h.
- Repos → Community +119d lead GitHub Trending Python first · then Show HN 12d agovolcengine/OpenViking
- Papers → Community +22d lead arXiv cs.AI first · then Hacker News front 13d agoMemorilla: Latent Semantic Memory for LLMs
- Papers → Community +14h lead arXiv cs.AI first · then Hacker News front 18d agoProving and Improving the Platonic Representation Hypothesis with JEPAs
- 1Rebuilding my local voice assistant pipelineblog.platypush.techLive Signal Community ·Practice · 2h old · seen in 2 sources HEATING ×2
- 2PauseAI Munich Rally | Stop the loss of control over AIlesswrong.comLive Signal Community ·Policy & safety · 18m old · via LessWrong
- 3We need commitments for 25% compute for safety by 205 ECI pointslesswrong.comLive Signal Community ·Policy & safety · 30m old · via LessWrong
- 4Show HN: Billet – 6MB (Agentic) Task Management Systemmktwsk.xyzLive Signal Community ·Tools · 54m old · via Show HN
- 5Show HN: A SEO Tool to get found by Google and get cited by AImozolite.comLive Signal Community ·Tools · 1h old · via Show HN
- 6Show HN: Fizzler – a simple tool that pulls agents out of defaultthinkfizzler.aiLive Signal Community ·Tools · 1h old · via Show HN
- 7Show HN: Taskthread – a task board your AI agents take work from (MCP)taskthread.denegra.aiLive Signal Community ·Tools · 1h old · via Show HN
- 8Frontier Clock: Tracking the US-China AI Gaplesswrong.comLive Signal Community ·Policy & safety · 2h old · via LessWrong
- 9Show HN: AdversaryPR – A Claude skill to keep up with AI-written PR'sgithub.comLive Signal Community ·Tools · 2h old · via Show HN
- 10Show HN: Dynobox – Integration tests for agents using CLI and MCP mocksgithub.comLive Signal Community ·Tools · 3h old · via Show HN
- 11Show HN: Daksha – personal assistant that remembers your life, cites sourcesdaksha.lifeLive Signal Community ·Tools · 3h old · via Show HN
- 12Show HN: Company docs didn't help our AI agent. Four shortnotes doubled accuracycontextflo.comLive Signal Community ·Practice · 3h old · via Show HN
- 13Show HN: PublicRows – pay-per-row data actors built from JSON recipespublicrows.comLive Signal Community · 3h old · via Show HN
- 14Show HN: Relic – An AI that only answers from what you've savedrelic-app.comLive Signal Community ·Tools · 3h old · via Show HN
- 15Show HN: Lintsure – Free local code check with rules from OWASP, EU AI Actgithub.comLive Signal Community ·Tools · 3h old · via Show HN
- 16Does most AI x-risk come from mostly-aligned models?lesswrong.comLive Signal Community ·Policy & safety · 4h old · via LessWrong
- 17Should governments slow frontier AI? Jacob Coxon and Robin Hanson debate [video]youtube.comLive Signal Community ·Policy & safety · 4h old · via Hacker News front
- 18Show HN: Eigin Pays Contributors for Specscontribute.eigin.aiLive Signal Community ·Business · 4h old · via Show HN
- 19Show HN: MCP-Sentry v2.0 – Zero-dependency firewall and audit proxy for MCPgithub.comLive Signal Community ·Tools · 4h old · via Show HN
- 20Claude solves two open problems in the physics of black holesanthropic.comLive Signal Community ·Research · 5h old · seen in 2 sources
- 21Show HN: Search your Google Maps saved placesmaptools.appLive Signal Community · 48m old · via Show HN
- 22Show HN: Lava Puzzle – Daily puzzle game about combining lava and waterlavapuzzle.comLive Signal Community · 1h old · via Show HN
- 23Show HN: Brisk – a lightweight, PyCharm-inspired editor written in Rustgithub.comLive Signal Community · 1h old · via Show HN
- 24Show HN: Wikipedia Current Event Portal as an Actual Newspaper (and in RSS)aeonfr.github.ioLive Signal Community · 2h old · via Show HN
- 25Show HN: TIE Fighter co-op in 11 days, built from 11 years of Wing Commanderfireflyk64.github.ioLive Signal Community · 2h old · via Show HN
- 26Show HN: Postdate – share a link now that opens at a time you choosepostdate.ioLive Signal Community · 2h old · via Show HN
- 27
- 28Show HN: IT Thinks – Build a GPU from a Single NAND Gatestore.steampowered.comLive Signal Community · 3h old · via Show HN
- 29Show HN: LZ0 -An asymmetric LZ compressor with ultra minimal, 0-alloc decodergithub.comLive Signal Community · 3h old · via Show HN
- 30Show HN: Plazzle – daily pathfinding puzzle gameplazzle.netlify.appLive Signal Community · 3h old · via Show HN
- 31
- 32A Critique of Dean Ball's "The Silent Consensus"lesswrong.comLive Signal Community · 3h old · via LessWrong
- 33OpenRun v1.0 - declarative deployments for internal tools on Docker or Kubernetesopenrun.devLive Signal Community · 4h old · via Lobsters
- 34Show HN: Roman Times – a Roman strategy game, playable in Latinromantimes.euLive Signal Community · 4h old · via Show HN
- 35Supporting FOSS projects to transition to memory safe techtweedegolf.nlLive Signal Community · 4h old · via Lobsters
- 36Mercurius — A Zero-Trust Network‑Native Window Systemmercurius.tebibyte.orgLive Signal Community · 4h old · via Lobsters
- 37The 98% Slot Machine: How Paid Advertising Became an Extraction Taxglobalcommercegraph.comLive Signal Community · 4h old · via Hacker News front
- 38
- 39Show HN: Orbital – Everything (known) in Earth orbit, live in 3D, in the browserjerredcook.github.ioLive Signal Community · 4h old · via Show HN
- 40
How Radar works
The gradient. Stories travel through the AI ecosystem in a rough sequence: they appear first in papers (arXiv, Hugging Face) or repos (GitHub), then surface in community discussion (Hacker News, Lobsters, LessWrong), then get interpreted in analysis (operator newsletters and essays). Radar ranks by where a story sits on that gradient. The earlier it is caught, the more lead time you have before it reaches consensus. The spectrum at the top of the page shows the last 24 hours of that flow — every story that entered the watched feeds, counted at the station where it first appeared.
The score. Each story gets a recency-decayed signal score. Every source tier carries a weight and a half-life, so live signal moves fast and fades fast while analysis sources move slowly and stay relevant longer. A story scores higher when two or more independent source families carry it — corroboration, not popularity, which is why five feeds of the same site count once. A small bonus applies when a story has visibly climbed the gradient. Stories past their tier's maximum age expire from the board.
Heating and lead time. A story is heating when it has surfaced in two or more independent source families. When the same story is observed first at an earlier stage and later at a higher one, Radar records the gap in hours — for example, seen on arXiv fourteen hours before it reached the Hacker News front page. That gap is the reader's lead time, and it is a measured span, never an estimate.
The classifier. Two judgments on this page are made by a model, not read from the feed log. A small typed classifier (TypeSafe's Jev) answers one question per story — is it about AI? — and stories it judges not to be about AI move to the bottom of the board; nothing is removed. It was checked against 232 hand-labelled titles before it shipped and agreed with 97.8% of them. The same classifier assigns the topic shown on a story and used by the filter above; a topic appears only when the classifier is confident, and at that setting it matched the labels on 115 of 116 stories. Stories without a confident topic are listed under All.
The honesty rule. Every other value on this page is an observed fact from the feed log — which source carried a story, when, and how many. Beyond those two labelled judgments, Radar makes no claim about what a story means; that is the newsletter's job. When a reading does not exist yet — no stories heating, nothing upstream — the page says so rather than inventing one.
- The spectrum band
- A filled flame tick is an early, lead-time-rich stage; a hollow tick is a later stage; a hollow ring on an empty span is a station with nothing on it yet.
- Signal class
- The tier chip: Live Signal moves fast and decays fast; analysis sources stay relevant longer. The ranker weights each accordingly.
- heating
- The same story in two or more independent source families — not two feeds of one site.
- seen in / via
- How many families carried it, or the single source that did.
Once a week, the operator read on what actually changed and the move it calls for. No digest, no hype.
Read the newsletter →