Hacker News

Article URL: https://www.science.org/content/article/how-ph-d-riding-bike

Comments URL: https://news.ycombinator.com/item?id=48433369

Points: 3

# Comments: 0

Article URL: https://ia601502.us.archive.org/23/items/pa-rating-area-individual-analyses-june-2026/PA_Rating_Area_Individual_Analyses_June2026.pdf

Comments URL: https://news.ycombinator.com/item?id=48433364

Points: 2

# Comments: 0

It's not X, it's Y.

That is so a genre or idiom of the current AI age. Why? I think because AIs live in language. They don't live somewhere else. So they have to exclude all the language network of things it's not in order to focus themselves on the thing it is. If they were more subconscious/symbolic like us, they could just not say that. But because their words, create their words, etc - more than us - they have to use this construction, to exclude the gigantic amount of stuff that whatever you're talking about is not.

What do y'all think?

Comments URL: https://news.ycombinator.com/item?id=48433363

Points: 2

# Comments: 1

Article URL: https://discourse.haskell.org/t/h2jvm-a-haskell-library-for-writing-jvm-bytecode/14182

Comments URL: https://news.ycombinator.com/item?id=48433356

Points: 2

# Comments: 0

Article URL: https://antirez.com/news/168

Comments URL: https://news.ycombinator.com/item?id=48433351

Points: 3

# Comments: 0

This is a project that I have been building for a while now, YourMemory is a solution to agentic memory which focuses on pruning of noise rather than hoarding of data.

In the current state of agentic memory most of the context is stored in the form of a MD file or is derived through a RAG model where you store each and everything. Both of the solution leads to bloated context which does not optimize the usage of any tokens.

In this system we only keep relevant data in our memory and prune all the unnecessary data. The relevance of a data is derived through multiple factors such as recall rate, importance, category, to which memory chain it's connected to etc. These parameters are fine tuned so that we can cater to both episodic memory and semantic memory.

Our memory layer keeps the size flat in this manner. You can draw correlation of this infrastructure with how Human brain store and prune memory.

The enterprise model is something very exciting as we can extract relevant memories from each user, agent and sub agent in this layer and that can be used by any one in the org, ensuring memory optimization at an enterprise level.

Comments URL: https://news.ycombinator.com/item?id=48433327

Points: 19

# Comments: 0

Article URL: https://plumbum.readthedocs.io/en/latest/

Comments URL: https://news.ycombinator.com/item?id=48433319

Points: 1

# Comments: 0

Article URL: https://tumbleweedwords.com/from-stamps-to-submittable-what-published-means-now.html

Comments URL: https://news.ycombinator.com/item?id=48433314

Points: 1

# Comments: 0

Article URL: https://arxiv.org/abs/2606.04299

Comments URL: https://news.ycombinator.com/item?id=48433305

Points: 1

# Comments: 0

Article URL: https://www.epc.eu/publication/The-strategic-importance-of-Europe-37f244/

Comments URL: https://news.ycombinator.com/item?id=48433304

Points: 1

# Comments: 0

Article URL: https://www.businessinsider.com/klarna-cmo-built-ai-replica-himself-colleagues-vent-at-2026-5

Comments URL: https://news.ycombinator.com/item?id=48433292

Points: 1

# Comments: 0

Article URL: https://github.com/loverofpizzas/heavybad

Comments URL: https://news.ycombinator.com/item?id=48433284

Points: 1

# Comments: 0

Hi folks, I found this setup on consummer hardware that seems to have great results on local hardware. - qwen 3.6 q6 - 450 K context using turboquant turbo3 mode llama.cpp fork - multimodal support

This AI generated blog article is a kind of "report" of what and how I did and result exemples.

I hope this can be usefull to some peopole.

Note : I am not much intersted in having success with this article, I mainly want to share what I think is an interesting use of a 5090. I generated the blog page telling AI to be compliant with hn "rules" and remain factual.

It's definitely not perfect, done rather quickly, not properly tested over 265K context. please forgive my lazyness :) . I am just enthousiast right now about what can be done on a 5090.

Comments URL: https://news.ycombinator.com/item?id=48433268

Points: 1

# Comments: 0

Article URL: https://www.hiig.de/wp-content/uploads/2020/03/twentyforty-book-1.pdf

Comments URL: https://news.ycombinator.com/item?id=48433267

Points: 1

# Comments: 0

Article URL: https://www.ft.com/content/8559a3f9-86de-4a1c-8a75-6623e83e6a00

Comments URL: https://news.ycombinator.com/item?id=48433260

Points: 2

# Comments: 0

Article URL: https://www.nytimes.com/2026/06/07/business/caastle-fraud-christine-hunsicker.html

Comments URL: https://news.ycombinator.com/item?id=48433255

Points: 1

# Comments: 0

Article URL: https://www.machinebrief.com/news/ai-memory-fails-spectacularly-95percent-error-rate-unveiled-ewj0

Comments URL: https://news.ycombinator.com/item?id=48433243

Points: 2

# Comments: 0

Article URL: https://status.sr.ht/issues/2026-06-06-llms-again/

Comments URL: https://news.ycombinator.com/item?id=48433196

Points: 4

# Comments: 0

Article URL: https://a-dark-cave.com/

Comments URL: https://news.ycombinator.com/item?id=48433186

Points: 1

# Comments: 1

Given how quickly things evolve, it's easy to get lost in the numerous offerings and hard to get the best deal. So, what do you use? Both clients/harnesses and LLM providers or local setups would be interesting.

Personally, I've been using opencode with Github copilot for work. I'm currently looking for cost-effective provider for personal work. Maybe openrouter with one of the cheap models?

Comments URL: https://news.ycombinator.com/item?id=48433171

Points: 1

# Comments: 0