Hacker News: loehnsberg

New comment by loehnsberg in "Show HN: Komi-learn – continuous memory and self-improvement for coding agents"

loehnsberg — Sun, 31 May 2026 06:48:55 +0000

It sounds like it solves the problem that everybody who vibe codes over multiple projects runs into, but it does not provide evidence that it actually works and that it is better than structured collection of md files. I think what is lacking in this field are benchmarks like LoCoMo for long sessions.

New comment by loehnsberg in "Google changes its search box"

loehnsberg — Tue, 19 May 2026 19:08:11 +0000

Time for Kagi MCP to become available to subscribers!

[1] https://github.com/kagisearch/kagimcp

New comment by loehnsberg in "Boris Cherny: TI-83 Plus Basic Programming Tutorial (2004)"

loehnsberg — Thu, 07 May 2026 12:27:21 +0000

Do you think Boris can still be reached under pickledcherry668@yahoo.com ?

New comment by loehnsberg in "DeepSeek V4 – almost on the frontier"

loehnsberg — Sat, 02 May 2026 20:48:39 +0000

Among the inexpensive models (and I include Grok 4.3 in this list), GLM 5.1 really sticks out!

On my personal test bench, when compared to other inexpensive models, GLM 5.1 provides the answers that I would consider most complete or satisfying (these are subjects that I consider myself an expert in). The answers tend to be more comprehensive, nuanced, and include references that I would consider the correct ones (if given access to web search).

I also find it a joy to code with, somewhere between Sonnet 4.6 and Opus 4.6 (have not tested Opus 4.7 yet).

Finally, just gauging by pelicans, it kind of stick out: https://simonwillison.net/tags/pelican-riding-a-bicycle/

New comment by loehnsberg in "Claude Brain"

loehnsberg — Sun, 19 Apr 2026 19:17:35 +0000

How does this work under the hood? What is so different from the OpenClaw approach of being able go do a semantic search over past sessions?

New comment by loehnsberg in "OpenClaw’s memory is unreliable, and you don’t know when it will break"

loehnsberg — Sat, 11 Apr 2026 09:17:30 +0000

I think if we want to build on what we have, instead of compaction at the end of the context window, the LLM would have to 'sleep', i.e. adjust its weights, then wake up with the last bits of the old context window in the new one, and have a 'feel' for what it did before through the change in weights. I just sense it's not that simple to get there, because simply updating the weights based on a single context sample risks degrading the weights of the whole network.

I like the idea of using small local model (or several) for tackling this problem, like low rank adaptation, but with current tech, I still have to piece this together or the small local models will forget old memories.

New comment by loehnsberg in "OpenClaw’s memory is unreliable, and you don’t know when it will break"

loehnsberg — Sat, 11 Apr 2026 07:50:24 +0000

As long as there's no solution to the long-term memory problem, we will have a "country of geniuses in a data center" that are all suffering from anterograde amnesia (movie: Memento), which requires human hand-holding.

I have experimented with a lot of hacks, like hierarchies of indexed md files, semantic DBs, embeddings, dynamic context retrieval, but none of this is really a comprehensive solution to get something that feels as intelligent as what these systems are able to do within their context windows.

I am als a touch skeptical that adjusting weights to learn context will do the trick without a transformer-like innovation in reinforcement learning.

Anyway, I‘ll keep tinkering…

New comment by loehnsberg in "European alternatives to Google, Apple, Dropbox and 120 US apps"

loehnsberg — Fri, 03 Apr 2026 11:20:34 +0000

I guess you can make that same argument about USA and America. Canada is clearly America but a Canadian would not refer to himself as American whereas a US national would. Europeans hardly refer to themselves as such but when European countries are lumped together, it has become common to ignore geography and refer to those affiliated with EU membership or bilateral EU affiliation (Norway, Switzerland, Iceland) as European.

New comment by loehnsberg in "My Random Forest Was Mostly Learning Time-to-Expiry Noise"

loehnsberg — Fri, 20 Mar 2026 04:57:05 +0000

The blogger trained a random forest on the returns of some price with a set of features and the ttm feature was strongest. Did I miss anything?

New comment by loehnsberg in "Hegseth gives Anthropic until Friday to back down on AI safeguards"

loehnsberg — Tue, 24 Feb 2026 22:39:21 +0000

I really hope they continue to show some spine against this administration and do not allow to weaponize AI against human beings. It's the morally right thing to do!

New comment by loehnsberg in "European Alternatives"

loehnsberg — Fri, 23 Jan 2026 16:02:33 +0000

Isn't it sad that we now have Russian, Chinese, American, European, etc alternatives? I mean I get it, Sept 11 paved the way for FISA orders and NSA overreach, Russia and China reverted back into dictatorship, but Europe is also at the edge. Shouldn't we rather fight that nationalistic power grab that just makes us all poorer and less free? And instead propagate global alternatives that are not subjected by some power-hungry state-/capital-sponsored overlord?

New comment by loehnsberg in "American importers and consumers bear the cost of 2025 tariffs: analysis"

loehnsberg — Mon, 19 Jan 2026 17:22:06 +0000

This is a perfect example of how to lie with statistics. All of these countries are either tax havens or oil-rich economies, apart from half of them having the population of a small city. The economic policy implemented by any of these countries cannot be implemented by a large economy with little or no natural resources, or would you recommend to Germany or Japan to just "HAVE" oil or open their banks as offshore foreign accounts?

New comment by loehnsberg in "I replaced Windows with Linux and everything's going great"

loehnsberg — Sat, 10 Jan 2026 17:23:15 +0000

Why Bedrock? Get the kids a Steam Deck, Prism Launcher, open a local server and boom :) It‘s not iOS convenience but they‘ll sure love to tinker with all the mods you can install.

New comment by loehnsberg in "AI Slop Report: The Global Rise of Low-Quality AI Videos"

loehnsberg — Sun, 28 Dec 2025 10:30:44 +0000

Haha… no bot here. Been using Kagi for years now. Not sure what you‘re searching for. My own tests, admittedly early, found no instances, where Google gave better results.

So You Want Continuous Time Zones (2017)

loehnsberg — Sat, 25 Oct 2025 11:47:54 +0000

Article URL: https://qntm.org/continuous

Comments URL: https://news.ycombinator.com/item?id=45703110

Points: 1

# Comments: 1

New comment by loehnsberg in "Fastmail desktop app"

loehnsberg — Mon, 13 Oct 2025 07:10:42 +0000

Indeed what I do as well; gives me apps for Youtube, Netflix, etc. The only downside is that you have to login if you do not use the "app" for a while. Would Electron get around this?

New comment by loehnsberg in "Starship's eleventh flight test"

loehnsberg — Sat, 04 Oct 2025 07:18:36 +0000

Suppose you have a liberal mindset and work there, you must bend the knee and practice anticipatory obedience, or why else would you tell the world that the rocket will be ”dropping into the Gulf of America?“

New comment by loehnsberg in "The best YouTube downloaders, and how Google silenced the press"

loehnsberg — Sat, 20 Sep 2025 11:13:35 +0000

It‘s still a stark abuse of power and borderline extortion by Google to use a private sentencing mechanism rather than dragging the purpetrator to public court over advertising and/or encouraging criminal activity, which may or may not have happened if Google Ads and Youtube were not part of the same monopolistic entity.

New comment by loehnsberg in "DeepSeek writes less secure code for groups China disfavors?"

loehnsberg — Wed, 17 Sep 2025 19:46:25 +0000

Did they use the online Deepseek Chat or the open source model. If you ask either about the Tianenmen Square you get very different answers, which may be true for response quality as well.

New comment by loehnsberg in "LLM leaderboard – Comparing models from OpenAI, Google, DeepSeek and others"

loehnsberg — Fri, 01 Aug 2025 08:43:44 +0000

Interesting to learn that o4-mini-high has the highest intelligence/$ score here at par with o3-pro which is twice as expensive and slow.