Harper Reed
Everything, newest first — across every channel. Their profile → · Subscribe
Hiding
21 October
20 October
17 October
16 October
15 October
- GitHubdotfiles — Dotfiles. Managed by YADM
- x.com
14 October
13 October
11 October
9 October
8 October
7 October
6 October
5 October
4 October
- x.com
- GitHubharper.micro.blog
3 October
1 October
28 September
27 September
24 September
22 September
From one piece Why Don't My Agents Break Containment? 3 beliefs, in the piece's order there
-
Their words
This type of experience must be part of a lot of these LLMs training. They are very effective at attacking these types of problems. They don't give up once it appears impossible, they just keep trying to figure out how to solve it.
-
Their words
if the llm thinks it is in some eval type of situation it will have a very different safety posture
-
Their words
Most of the defensive work we do building agentic systems is directly rooted in token scarcity.
21 September
20 September
19 September
11 September
9 September
8 September
2 September
1 September
28 August
18 August
17 August
16 August
15 August
From one piece Notes From a Fugue State: Put the Things in Your Head That You Love 3 beliefs, in the piece's order there
-
korrents.com
The companies that win with AI will be those that use it to do far more, not merely to operate more efficiently.Their words
We will look back at this time period and laugh at all the companies who are using AI to make things MUCH MORE EFFICIENT. They will seem lazy and boring. The winners will be the companies that do so much more than they are currently doing.
-
Their words
There will be a correction once the CFOs realize that a junior that costs $120k is equivalent to a mid stage employee who costs $350k.
-
Their words
you can't make a bad candidate better with superior technology. Elections are won by the candidates, not the tech stack.
14 August
9 August
From one piece How to find magical travel experiences using an AI travel buddy 2 beliefs, in the piece's order there
-
korrents.com
AI assistants still make mistakes and their answers need to be double-checked by a human.Their words
This thing still gets things wrong, and you really have to double check.
-
Their words
It has gotten really good over the last year or so. Especially in places where I don't speak the language.
16 July
11 June
18 May
14 May
11 March
From one piece My now immaculate knowledge graph of life 3 beliefs, in the piece's order there
-
Their words
the best way to build a skill is to start doing the work manully with claude, and the to tell claude to build a skill out the previous few rounds of work.
-
Their words
Your interface should feel more like Claude than a traditional file browser or note-taking app.
-
Their words
Do not waste energy forcing everything into some perfect predefined system. Put things where they fall, not where some abstract framework says they should go. The system should follow your work, not the other way around.
8 January
-
Their words
I do not think that 2026 is going to bring a reprieve from the horrors. Only amplify.
6 January
-
korrents.com
Early-2000s software development tooling was mostly bad, despite nostalgia for that era.Their words
There is a lot of nostalgia for that time period - but for the most part the tooling sucked.
3 December 2025
From one piece Getting Claude Code to do my emails 2 beliefs, in the piece's order there
-
korrents.com
Giving AI agents the ability to take actions that affect other people is risky and should be approached with caution.Their words
Giving your agents access to things that affect other people is scary and should be done with caution.
-
Their words
This is obviously the future. Whether we like it or not, it appears that agentic email will be a thing.
30 September 2025
From one piece We Gave Our AI Agents Twitter and Now They're Demanding Lambos 3 beliefs, in the piece's order there
-
Their words
It seems that the agents want human tools. We know this cuz we give agents human tools all the time within the codegen tooling: git, ls, readfile, writefile, cat, etc.
-
Their words
I think this is because the models are overweighted with content from this time period.
-
Their words
My pet theory is that we are speed-running early 2000s enterprise software development lifecycle and work style.
8 May 2025
17 April 2025
10 April 2025
16 February 2025
13 January 2025
1 January 2025
4 December 2024
6 November 2024
31 July 2024
12 April 2024
26 March 2024
18 March 2024
Nothing matches.
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.