Tap a claim on the ring to put it at the centre.
← The code today’s models and agents write is very hard to follow: you…
17 connected korrents · 15 moments on record from 25 Aug 2025 to 3 Sept 2026.
Everything filed under Anthropic
Anthropic
Everything filed under software quality
software quality
Everything filed under coding agents
coding agents
Everything filed under MCP
MCP 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 code today’s models and agents write is very hard to follow: you get better performance without knowing why, and no way to fix it when it breaks. The code today’s models and agents write is very hard to follow: you get better performance without knowing why, and no way to fix it when it breaks. Last stated 2 months ago
19 Jul 2026
ES
Elizabeth Stone — holds since 2026-07-19 — tap for who they are
Same subject: What limits you with coding agents is your own skill at stringing them together, not the capability of the models. — tap to centre the map on it What limits you with coding agents is your own skill at stringing them together, not the capability of the models. Last stated 6 months ago
20 Mar 2026
AK
Andrej Karpathy — holds since 2026-03-20 — tap for who they are
Same subject: Coding models are worst at exactly the thing an AI research explosion would need: code that has never been written before. — tap to centre the map on it Coding models are worst at exactly the thing an AI research explosion would need: code that has never been written before. Last stated 11 months ago
17 Oct 2025
AK
Andrej Karpathy — holds since 2025-10-17 — tap for who they are
Same subject: Agent-written code that runs and passes its tests is not enough: security, maintainability and being able to roll it back still need humans in the loop. — tap to centre the map on it Agent-written code that runs and passes its tests is not enough: security, maintainability and being able to roll it back still need humans in the loop. Last stated 3 months ago
7 Jun 2026
TF
Tony Fadell — holds since 2026-06-07 — tap for who they are
Same subject: The way software gets built flipped in December: writing code yourself is now the exception, not the default. — tap to centre the map on it The way software gets built flipped in December: writing code yourself is now the exception, not the default. Last stated 6 months ago
20 Mar 2026
AK
Andrej Karpathy — holds since 2026-03-20 — tap for who they are
Same subject: Agentic code review raises the floor but cannot be trusted, because the model reading the code is the same model that wrote it, and it will tell you the code is great. — tap to centre the map on it Agentic code review raises the floor but cannot be trusted, because the model reading the code is the same model that wrote it, and it will tell you the code is great. Last stated 2 months ago
15 Jul 2026
DH
Dex Horthy — holds since 2026-07-15 — tap for who they are
Same subject: The verbose enterprise patterns everyone hated are worth having again: coding agents are idiots working around the clock, and you are not the one typing the boilerplate any more. — tap to centre the map on it The verbose enterprise patterns everyone hated are worth having again: coding agents are idiots working around the clock, and you are not the one typing the boilerplate any more. Last stated 3 months ago
27 May 2026
DR
Dax Raad — holds since 2026-05-27 — tap for who they are
Same subject: Long-running agents fail because they drift off the distribution they were trained on, and degrade further the farther out they get. — tap to centre the map on it Long-running agents fail because they drift off the distribution they were trained on, and degrade further the farther out they get. Last stated a month ago
30 Jul 2026
JD
Jeff Dean — holds since 2026-07-30 — tap for who they are
Same subject: Background agents do not work for real development, because steering a model as it drifts is the job, and you cannot steer what you are not watching. — tap to centre the map on it Background agents do not work for real development, because steering a model as it drifts is the job, and you cannot steer what you are not watching. Last stated a year ago
25 Aug 2025
PS
Peter Steinberger — holds since 2025-08-25 — tap for who they are
Same subject: A language-agnostic conformance suite is the most powerful thing you can hand a coding agent, because the whole instruction becomes: write code until these tests pass. — tap to centre the map on it A language-agnostic conformance suite is the most powerful thing you can hand a coding agent, because the whole instruction becomes: write code until these tests pass. Last stated 6 months ago
19 Mar 2026
SW
Simon Willison — holds since 2026-03-19 — tap for who they are
Same subject: A passing test suite is not evidence that the software works, so an agent must also be made to start the thing and exercise it the way a person would. — tap to centre the map on it A passing test suite is not evidence that the software works, so an agent must also be made to start the thing and exercise it the way a person would. Last stated 6 months ago
19 Mar 2026
SW
Simon Willison — holds since 2026-03-19 — tap for who they are
Same subject: Adding a config option so nobody breaks compounds into permutations no test suite can cover, because it is infinitely harder to evolve software that has users. — tap to centre the map on it Adding a config option so nobody breaks compounds into permutations no test suite can cover, because it is infinitely harder to evolve software that has users. Last stated 4 weeks ago
10 Aug 2026
PS
Peter Steinberger — holds since 2026-08-10 — tap for who they are
Same subject: A benchmark that ranks Claude Code last while it stays first in use is measuring the wrong thing, and has been for a year. — tap to centre the map on it A benchmark that ranks Claude Code last while it stays first in use is measuring the wrong thing, and has been for a year. Last stated 4 days ago
3 Sept 2026
DR
Dax Raad — holds since 2026-09-03 — tap for who they are
Same subject: A language model asked to summarize its own system prompt risks that prompt's content biasing the summary it produces. — tap to centre the map on it A language model asked to summarize its own system prompt risks that prompt's content biasing the summary it produces. Last stated 5 days ago
2 Sept 2026
SW
Simon Willison — holds since 2026-09-02 — tap for who they are
Same subject: A trust that owns the mission protects a company better than founder control does, which is why Anthropic needs no dual-class shares. — tap to centre the map on it A trust that owns the mission protects a company better than founder control does, which is why Anthropic needs no dual-class shares. Last stated 4 months ago
10 May 2026
ER
Eric Ries — holds since 2026-05-10 — tap for who they are
Same subject: Every tool is better as a CLI than as an MCP server: models are good at Unix commands, a CLI composes with jq, and MCP output clutters the context. — tap to centre the map on it Every tool is better as a CLI than as an MCP server: models are good at Unix commands, a CLI composes with jq, and MCP output clutters the context. Last stated 7 months ago
12 Feb 2026
PS
Peter Steinberger — holds since 2026-02-12 — tap for who they are
Same subject: MCP is not a new foundation for anything — it is API design, and Kubernetes was already turning one intention into seven calls a decade ago. — tap to centre the map on it MCP is not a new foundation for anything — it is API design, and Kubernetes was already turning one intention into seven calls a decade ago. Last stated 3 months ago
3 Jun 2026
KH
Kelsey Hightower — holds since 2026-06-03 — tap for who they are
Same subject: MCP servers are worth removing outright: an agent reading the code directly is faster, and pollutes its context less, than one that can reach for a tool unasked. — tap to centre the map on it MCP servers are worth removing outright: an agent reading the code directly is faster, and pollutes its context less, than one that can reach for a tool unasked. Last stated a year ago
25 Aug 2025
PS
Peter Steinberger — holds since 2025-08-25 — tap for who they are
At the centre
The code today’s models and agents write is very hard to follow: you get better performance without knowing why, and no way to fix it when it breaks.
Holds ES Elizabeth StoneRead this korrent →
What limits you with coding agents is your own skill at stringing them together, not the capability of the models.
Holds
Coding models are worst at exactly the thing an AI research explosion would need: code that has never been written before.
Holds
Agent-written code that runs and passes its tests is not enough: security, maintainability and being able to roll it back still need humans in the loop.
Holds TF Tony Fadell
The way software gets built flipped in December: writing code yourself is now the exception, not the default.
Holds
Agentic code review raises the floor but cannot be trusted, because the model reading the code is the same model that wrote it, and it will tell you the code is great.
Holds DH Dex Horthy
The verbose enterprise patterns everyone hated are worth having again: coding agents are idiots working around the clock, and you are not the one typing the boilerplate any more.
Holds
Long-running agents fail because they drift off the distribution they were trained on, and degrade further the farther out they get.
Holds JD Jeff Dean
Background agents do not work for real development, because steering a model as it drifts is the job, and you cannot steer what you are not watching.
Holds
A language-agnostic conformance suite is the most powerful thing you can hand a coding agent, because the whole instruction becomes: write code until these tests pass.
Holds
A passing test suite is not evidence that the software works, so an agent must also be made to start the thing and exercise it the way a person would.
Holds
Adding a config option so nobody breaks compounds into permutations no test suite can cover, because it is infinitely harder to evolve software that has users.
Holds
A benchmark that ranks Claude Code last while it stays first in use is measuring the wrong thing, and has been for a year.
Holds
A language model asked to summarize its own system prompt risks that prompt's content biasing the summary it produces.
Holds
A trust that owns the mission protects a company better than founder control does, which is why Anthropic needs no dual-class shares.
Holds ER Eric Ries
Every tool is better as a CLI than as an MCP server: models are good at Unix commands, a CLI composes with jq, and MCP output clutters the context.
Holds
MCP is not a new foundation for anything — it is API design, and Kubernetes was already turning one intention into seven calls a decade ago.
Holds KH Kelsey Hightower
MCP servers are worth removing outright: an agent reading the code directly is faster, and pollutes its context less, than one that can reach for a tool unasked.
Holds