Live data from Hacker News

Viewing profile — langs

langs

HN member
Joined
Wed, Jan 18, 2012, 11:12 AM UTC
HN karma
28
Public activity
21 items

About langs

No profile information was provided.

Recent public activity

  1. comment
    Comment #49238373

    Simply feeding the Slay the Spire game state directly to an LLM is currently insufficient to achieve stable deck-building or consistent combat performance. I categorize the game's …

  2. comment
    Comment #49237960

    I'm building an AI agent that plays Slay the Spire. It currently reaches A20 Act 3 consistently and occasionally defeats the A20 Heart. It use two search: spare graph search for de…

  3. comment
    Comment #49180519

    Yes, it uses modern technology to tackle an old problem: retrieval.

  4. comment
    Comment #49179835

    https://github.com/AttemorySystem/attemory/ stars and issues are welcome :) Using attention for retrieval was inspired by a comment I saw in here long time ago: Prediction and retr…

  5. comment
    Comment #49179349

    I am working on the same thing right now. However, unlike storing conversations in an external retrieval system, I use a local LLM to store the conversation's KV cache and perform …

  6. comment
    Comment #49177889

    Pi’s minimalism reveals a simple truth: the LLM is the core of any agent harness. Consequently, much of current harness tuning will become redundant(or even a hindrance) with next-…

  7. comment
    Comment #49177827

    > Myth 2: Writing Code Is the Bottleneck Writing code is indeed the bottleneck for same resource constrained companies. Rapid code development creates more opportunities for trial …

  8. comment
    Comment #49167148

    You need to optimize the KVCache part(save to disk to save compute) to achieve this goal.

  9. comment
    Comment #49120500

    So, is this a competition? To see whose model can be jailbroken the most times and incite the highest level of public alarm?

  10. story
  11. comment
    Comment #49078464

    "mandatory safety testing" is an impractical ideal, it's not workable in real world. Like any technology, LLMs are dual-use tools capable of both beneficial and malicious applicati…

  12. comment
    Comment #48929524

    the current bottlenecks for agents are actually bottlenecks of the LLMs, rather than the agents themselves.

  13. comment
    Comment #48902047

    That’s a good analogy, but I think the reason we currently need the "nurse" role is the need to interact with the physical world. Most software products don't require this step, so…

  14. comment
    Comment #48901664

    How does your solution differ from other KV Cache approaches, such as LMCache?

  15. comment
  16. comment
  17. comment
    Comment #48856856

    nope. any rewrite will be an AI rewrite soon.

  18. story
  19. comment
    Comment #47884945

    I vide-coded some, but the core is hand-craft arch/algorithm, that's the most valuable part I want to protect.

  20. comment
  21. story
    Ask HN: How do solo devs protect their work in the age of vibe coding?

    I am working on a new open-source project. (My project is in AI infrastructure. It already gets SOTA results on several well-known benchmarks.) The core value is not just the code,…