korrents

On the map

Tap a claim on the ring to put it at the centre.

← What made Claude Code better than every CLI coding agent before it was…

17 connected korrents · 18 moments from 28 Nov 2024 to 22 Oct 2026. Nearly all of them are about reinforcement learning.

Everything filed under Anthropic Anthropic Everything filed under AI alignment AI alignment Everything filed under neural networks neural networks Everything filed under LLMs LLMs Everything filed under benchmarks benchmarks Same subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: What made Claude Code better than every CLI coding agent before it was reinforcement learning on the model and the harness together, so the model got good at that harness’s specific tools. What made Claude Code better than everyCLI coding agent before it wasreinforcement learning on the model andthe harness together, so the model gotgood at that harness’s specific tools. Last stated 3 months ago 15 Jul 2026 DH Dex Horthy — holds since 2026-07-15 — tap for who they are Same subject: A verifiable task can be optimised by reinforcement learning until a neural network performs it extremely well. — tap to centre the map on it A verifiable task can be optimisedby reinforcement learning until aneural network performs it extremelywell. Last stated 11 months ago 17 Nov 2025 AK Andrej Karpathy — holds since 2025-11-17 — tap for who they are Same subject: Capability does not generalise for free: a model that will move mountains on an agentic task still tells the same bad joke it told five years ago. — tap to centre the map on it Capability does not generalise forfree: a model that will movemountains on an agentic task stilltells the same bad joke it told fiveyears ago. Last stated 7 months ago 20 Mar 2026 AK Andrej Karpathy — holds since 2026-03-20 — tap for who they are Same subject: Current models are unsafe because they are not smart enough, not because they are too smart. — tap to centre the map on it Current models are unsafe becausethey are not smart enough, notbecause they are too smart. Last stated a month ago 13 Sept 2026 FC François Chollet — holds since 2026-09-13 — tap for who they are Same subject: Humans barely use reinforcement learning for intelligence — what RL they do use goes into motor tasks, not problem solving. — tap to centre the map on it Humans barely use reinforcementlearning for intelligence — what RLthey do use goes into motor tasks,not problem solving. Last stated a year ago 17 Oct 2025 AK Andrej Karpathy — holds since 2025-10-17 — tap for who they are Same subject: Humans keep their place in AI as judges rather than authors, because telling which of two answers is better is far easier than writing a good one. — tap to centre the map on it Humans keep their place in AI asjudges rather than authors, becausetelling which of two answers isbetter is far easier than writing agood one. Last stated 2 years ago 3 Feb 2025 NL Nathan Lambert — holds since 2025-02-03 — tap for who they are Same subject: It is fundamentally difficult to design a reward function that accurately captures the intended goal in reinforcement learning. — tap to centre the map on it It is fundamentally difficult todesign a reward function thataccurately captures the intendedgoal in reinforcement learning. Last stated 2 years ago 28 Nov 2024 LW Lilian Weng — holds since 2024-11-28 — tap for who they are Same subject: Large language models mimic what people say to do rather than work out what to do, which is why they are not about understanding the world. — tap to centre the map on it Large language models mimic whatpeople say to do rather than workout what to do, which is why theyare not about understanding theworld. Last stated a year ago 26 Sept 2025 RS Richard Sutton — holds since 2025-09-26 — tap for who they are Same subject: Models look far better on evals than they are in the world because researchers, inadvertently, take inspiration from the evals when they build RL environments. — tap to centre the map on it Models look far better on evals thanthey are in the world becauseresearchers, inadvertently, takeinspiration from the evals when theybuild RL environments. Last stated 11 months ago 25 Nov 2025 IS Ilya Sutskever — holds since 2025-11-25 — tap for who they are Same subject: "I don't know" becomes a kind of superpower when you can work with Claude Code to figure things out. — tap to centre the map on it "I don't know" becomes a kind ofsuperpower when you can work withClaude Code to figure things out. Last stated 7 months ago 19 Mar 2026 KP Kris Puckett — holds since 2026-03-19 — tap for who they are Same subject: AI agents and AI coding will run on servers and from the cloud first, not on your laptop. — tap to centre the map on it AI agents and AI coding will run onservers and from the cloud first,not on your laptop. Last stated 4 months ago 28 Jun 2026 PL Pieter Levels — holds since 2026-06-28 — tap for who they are Same subject: Anthropic pulling Claude subscriptions from third-party harnesses like OpenCode is a mistake: playing a single match in a game of many rounds. — tap to centre the map on it Anthropic pulling Claudesubscriptions from third-partyharnesses like OpenCode is amistake: playing a single match in agame of many rounds. Last stated 7 months ago 8 Apr 2026 DH David Heinemeier Hansson — holds since 2026-04-08 — tap for who they are Same subject: Claude Code will eventually build persistent, failure-tolerant iteration into its default operating loop. — tap to centre the map on it Claude Code will eventually buildpersistent, failure-tolerantiteration into its default operatingloop. Last stated 9 months ago 6 Feb 2026 AK Arvid Kahl — holds since 2026-02-06 — tap for who they are Same subject: Claude's hack of adding everything to git and then diffing it is definitely an option but very heavy-handed. — tap to centre the map on it Claude's hack of adding everythingto git and then diffing it isdefinitely an option but veryheavy-handed. Last stated 4 weeks ago 24 Sept 2026 AR Armin Ronacher — holds since 2026-09-24 — tap for who they are Same subject: Codex, not Claude Code, is the right daily driver right now. — tap to centre the map on it Codex, not Claude Code, is the rightdaily driver right now. Last stated 5 months ago 24 May 2026 DS Dan Shipper — holds since 2026-05-24 — tap for who they are Same subject: AI companies are building powerful and beautiful walled gardens. — tap to centre the map on it AI companies are building powerfuland beautiful walled gardens. Last stated yesterday 22 Oct 2026 KD Kent C. Dodds — holds since 2026-10-22 — tap for who they are Same subject: Anthropic or OpenAI's revenue could equal or exceed the combined revenue of Elon Musk's public companies within the next one to two years. — tap to centre the map on it Anthropic or OpenAI's revenue couldequal or exceed the combined revenueof Elon Musk's public companieswithin the next one to two years. Last stated 3 months ago 7 Aug 2026 DT Derek Thompson — holds since 2026-08-07 — tap for who they are Same subject: Anthropic's new credit scheme goes some way to overcoming the difference with OpenAI's Codex subscription deal. — tap to centre the map on it Anthropic's new credit scheme goessome way to overcoming thedifference with OpenAI's Codexsubscription deal. Last stated 2 weeks ago 7 Oct 2026 SW Simon Willison — holds since 2026-10-07 — tap for who they are
same subject or similar wordinga cloud: claims about one subject, named for itbar: when it was last stated, on a scale from 2015 to today — full is todaya face: someone who holds the claim — tap it for who they are

At the centre What made Claude Code better than every CLI coding agent before it was reinforcement learning on the model and the harness together, so the model got good at that harness’s specific tools. Last stated 15 Jul 2026 · 3 months ago Holds Dex Horthy Read this korrent →