Tap a claim on the ring to put it at the centre.
← The usual trade-off between quality and cost does not apply to…
17 connected korrents · 15 moments on record from 29 May 2019 to 6 Sept 2026.
Everything filed under reinforcement learning
reinforcement learning
Everything filed under software performance
software performance
Everything filed under scaling laws
scaling laws
Everything filed under design
design
Everything filed under coding agents
coding agents Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Read this korrent: The usual trade-off between quality and cost does not apply to software: high internal quality is the cheaper way to produce it. The usual trade-off between quality and cost does not apply to software: high internal quality is the cheaper way to produce it. Last stated 7 years ago
29 May 2019
MF
Martin Fowler — holds since 2019-05-29 — tap for who they are
Same subject: Software's real advantage over physical engineering is consistency: a component does what it is specified to do every time, while physical materials only fall within a tolerance. — tap to centre the map on it Software's real advantage over physical engineering is consistency: a component does what it is specified to do every time, while physical materials only fall within a tolerance. Last stated 6 years ago
20 Jan 2021
HW
Hillel Wayne — holds since 2021-01-20 — tap for who they are
Same subject: There is no trade-off between well-architected code and fast code: in practice, the code that is architected properly is also the code that runs quickly. — tap to centre the map on it There is no trade-off between well-architected code and fast code: in practice, the code that is architected properly is also the code that runs quickly. Last stated 2 weeks ago
26 Aug 2026
CM
Casey Muratori — holds since 2026-08-26 — tap for who they are
Same subject: Software's constraints are soft where traditional engineering's are hard, and that is one of the few real differences between them. — tap to centre the map on it Software's constraints are soft where traditional engineering's are hard, and that is one of the few real differences between them. Last stated 6 years ago
20 Jan 2021
HW
Hillel Wayne — holds since 2021-01-20 — tap for who they are
Same subject: Cheaper software has increased the number of software engineers a company needs, not reduced it. — tap to centre the map on it Cheaper software has increased the number of software engineers a company needs, not reduced it. Last stated a month ago
28 Jul 2026
BS
Blake Scholl — holds since 2026-07-28 — tap for who they are
Same subject: RLHF is not a tax on capability: preference tuning also raises maths and code scores, which is why the labs keep reaching for it. — tap to centre the map on it RLHF is not a tax on capability: preference tuning also raises maths and code scores, which is why the labs keep reaching for it. Last stated 2 years ago
3 Feb 2025
NL
Nathan Lambert — holds since 2025-02-03 — tap for who they are
Same subject: Software is now won or lost on design: good enough is no longer fine, it is mediocre. — tap to centre the map on it Software is now won or lost on design: good enough is no longer fine, it is mediocre. Last stated 11 months ago
16 Oct 2025
DF
Dylan Field — holds since 2025-10-16 — tap for who they are
Same subject: Software only gets good by being used a lot by the people building it, so time spent inside coding agents instead of the app is time that makes the app worse. — tap to centre the map on it Software only gets good by being used a lot by the people building it, so time spent inside coding agents instead of the app is time that makes the app worse. Last stated yesterday
6 Sept 2026
BV
Ben Vinegar — holds since 2026-09-06 — tap for who they are
Same subject: Quality comes from doing irrational things — half of what you do you did not have to do — and almost nobody is willing to operate that way. — tap to centre the map on it Quality comes from doing irrational things — half of what you do you did not have to do — and almost nobody is willing to operate that way. Last stated 3 months ago
27 May 2026
DR
Dax Raad — holds since 2026-05-27 — 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 optimised by reinforcement learning until a neural network performs it extremely well. Last stated 10 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 for free: a model that will move mountains on an agentic task still tells the same bad joke it told five years ago. Last stated 6 months ago
20 Mar 2026
AK
Andrej Karpathy — holds since 2026-03-20 — 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 reinforcement learning for intelligence — what RL they do use goes into motor tasks, not problem solving. Last stated 11 months ago
17 Oct 2025
AK
Andrej Karpathy — holds since 2025-10-17 — tap for who they are
Same subject: "Is the page loaded?" was never a real question, and web performance only became measurable once it was broken into the distinct moments a user actually notices. — tap to centre the map on it "Is the page loaded?" was never a real question, and web performance only became measurable once it was broken into the distinct moments a user actually notices. Last stated 3 weeks ago
19 Aug 2026
AO
Addy Osmani — holds since 2026-08-19 — tap for who they are
Same subject: "Optimize it later" would be safe advice only if every engineer already knew how to avoid the architectural mistakes that no later optimization can remove. — tap to centre the map on it "Optimize it later" would be safe advice only if every engineer already knew how to avoid the architectural mistakes that no later optimization can remove. Last stated 2 weeks ago
26 Aug 2026
CM
Casey Muratori — holds since 2026-08-26 — tap for who they are
Same subject: A 300-millisecond response budget is not an achievement but an admission: 300 milliseconds is an eternity in computing, and a product pitching it shows how far the bar has fallen. — tap to centre the map on it A 300-millisecond response budget is not an achievement but an admission: 300 milliseconds is an eternity in computing, and a product pitching it shows how far the bar has fallen. Last stated 2 weeks ago
26 Aug 2026
CM
Casey Muratori — holds since 2026-08-26 — tap for who they are
Same subject: A benchmark result should be reported under a stated budget, or as a curve against test-time compute — never as a single number. — tap to centre the map on it A benchmark result should be reported under a stated budget, or as a curve against test-time compute — never as a single number. Last stated 2 months ago
26 Jun 2026
NB
Noam Brown — holds since 2026-06-26 — tap for who they are
Same subject: A lab should spend most of its compute on research rather than on building the next model, because research is where the tenfold yearly efficiency gains come from. — tap to centre the map on it A lab should spend most of its compute on research rather than on building the next model, because research is where the tenfold yearly efficiency gains come from. Last stated 6 months ago
13 Mar 2026
DP
Dylan Patel — holds since 2026-03-13 — tap for who they are
Same subject: After pre-training, post-training and test-time scaling, the fourth scaling law is agentic: multiplying AI by spawning agents, and the whole loop scales on one thing, compute. — tap to centre the map on it After pre-training, post-training and test-time scaling, the fourth scaling law is agentic: multiplying AI by spawning agents, and the whole loop scales on one thing, compute. Last stated 6 months ago
23 Mar 2026
JH
Jensen Huang — holds since 2026-03-23 — tap for who they are
At the centre
The usual trade-off between quality and cost does not apply to software: high internal quality is the cheaper way to produce it.
Holds MF Martin FowlerRead this korrent →
Software's real advantage over physical engineering is consistency: a component does what it is specified to do every time, while physical materials only fall within a tolerance.
Holds HW Hillel Wayne
There is no trade-off between well-architected code and fast code: in practice, the code that is architected properly is also the code that runs quickly.
Holds CM Casey Muratori
Software's constraints are soft where traditional engineering's are hard, and that is one of the few real differences between them.
Holds HW Hillel Wayne
Cheaper software has increased the number of software engineers a company needs, not reduced it.
Holds BS Blake Scholl
RLHF is not a tax on capability: preference tuning also raises maths and code scores, which is why the labs keep reaching for it.
Holds NL Nathan Lambert
Software is now won or lost on design: good enough is no longer fine, it is mediocre.
Holds DF Dylan Field
Software only gets good by being used a lot by the people building it, so time spent inside coding agents instead of the app is time that makes the app worse.
Holds
Quality comes from doing irrational things — half of what you do you did not have to do — and almost nobody is willing to operate that way.
Holds
A verifiable task can be optimised by reinforcement learning until a neural network performs it extremely well.
Holds
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.
Holds
Humans barely use reinforcement learning for intelligence — what RL they do use goes into motor tasks, not problem solving.
Holds
"Is the page loaded?" was never a real question, and web performance only became measurable once it was broken into the distinct moments a user actually notices.
Holds AO Addy Osmani
"Optimize it later" would be safe advice only if every engineer already knew how to avoid the architectural mistakes that no later optimization can remove.
Holds CM Casey Muratori
A 300-millisecond response budget is not an achievement but an admission: 300 milliseconds is an eternity in computing, and a product pitching it shows how far the bar has fallen.
Holds CM Casey Muratori
A benchmark result should be reported under a stated budget, or as a curve against test-time compute — never as a single number.
Holds NB Noam Brown
A lab should spend most of its compute on research rather than on building the next model, because research is where the tenfold yearly efficiency gains come from.
Holds DP Dylan Patel
After pre-training, post-training and test-time scaling, the fourth scaling law is agentic: multiplying AI by spawning agents, and the whole loop scales on one thing, compute.
Holds JH Jensen Huang