Hacker News: sovenyr

New comment by sovenyr in "Agents for financial services and insurance"

sovenyr — Wed, 06 May 2026 04:53:43 +0000

this is to risky as for me!

New comment by sovenyr in "Agents can now create Cloudflare accounts, buy domains, and deploy"

sovenyr — Wed, 06 May 2026 04:52:39 +0000

don't even supricezed - I've done it before even without agents

New comment by sovenyr in "Ask HN: Best Embedding Models?"

sovenyr — Wed, 06 May 2026 04:14:16 +0000

please check OpenAI embedding models - especially small one

New comment by sovenyr in "Ask HN: Do you think AI will end up like dotcom?"

sovenyr — Thu, 30 Apr 2026 03:06:52 +0000

pretty much - true

New comment by sovenyr in "Ask HN: Can HN ban new accounts? or charge money?"

sovenyr — Wed, 29 Apr 2026 10:58:05 +0000

it should be better bot detection system - especially because of ai generated comments - and that's all

New comment by sovenyr in "Ask HN: Is it just me or is Claude Code getting worse?"

sovenyr — Tue, 28 Apr 2026 16:31:53 +0000

can't say it become worse, but at some point it stops be so useful as it was before - it looks like magic disapear

New comment by sovenyr in "Sabotaging projects by overthinking, scope creep, and structural diffing"

sovenyr — Sat, 25 Apr 2026 19:52:28 +0000

The flip side is that "fulfil your vision" requires actually having a vision before each session. Without explicit scope set upfront, the LLM defaults to filling time with adjacent improvements that feel like progress.

  I've started writing a one-line success criterion before opening the editor, the way @sambaumann's parent comment hints at — "the failure-mode-resistant version of this task is X." Not because the LLM can't figure out what I want, but because if I can't compress the goal to one        
  sentence, I'm not actually ready to build, I'm ready to wander.                                                                                                                                                                                                                               
                                                                                                                                                                                                                                                                                                
  The scope creep cure pre-LLM was fatigue. Now fatigue takes much longer to kick in, so the cure has to be cognitive instead.

New comment by sovenyr in "Google plans to invest up to $40B in Anthropic"

sovenyr — Sat, 25 Apr 2026 19:49:04 +0000

  The money does leave the room — it gets converted into physical infrastructure that's increasingly rationed. Capital is not the constraint anymore.                                                                                                                                           
                                                                                                                                                                                                                                                                                                
  Dominion Energy can't serve new data center load in Loudoun County until 2028. NERC flagged grid emergency risk across most of CONUS. Phoenix is in active water rationing while hosting 47 hyperscale facilities. Helium supply (45% Qatar, used in EUV cooling) took real hits in 2026.     
                                                                                                                                                                                                                                                                                                
  Even if Google funds Anthropic, who funds the new transformers, the grid interconnect queue position, the water permit reviews? That part of the cycle isn't circular — it's hitting physics.                                                                                                 
                  
  Capital is fungible. Substations aren't.