Hacker News: lysecret

New comment by lysecret in "I’ve joined Anthropic"

lysecret — Tue, 19 May 2026 17:23:17 +0000

Honestly happy he’s back at a foundation lab. He will have insane impact there. Of course one of the best educators in the world it’s a bit sad he gave up building and education tool.

New comment by lysecret in "I don't think AI will make your processes go faster"

lysecret — Sun, 17 May 2026 16:40:22 +0000

Ai in my mind is a new primitive of computing, like compute a db blob storage.

Goblin Hunt

lysecret — Sun, 10 May 2026 18:33:27 +0000

Article URL: https://blog.leloew.com/goblin-hunt/

Comments URL: https://news.ycombinator.com/item?id=48086516

Points: 6

# Comments: 0

New comment by lysecret in "A recent experience with ChatGPT 5.5 Pro"

lysecret — Sat, 09 May 2026 08:18:42 +0000

There is a great recent episode of latent space about a similar topic it’s worth a watch even with the click baiti thumbnail and title https://youtu.be/9d899Ram9Bs?is=pQMoVmlWVsTNKfRK

New comment by lysecret in "Claude Opus 4.7"

lysecret — Thu, 16 Apr 2026 17:30:59 +0000

What’s the default context window? Seems extremely short.

New comment by lysecret in "If DSPy is so great, why isn't anyone using it?"

lysecret — Mon, 23 Mar 2026 15:46:06 +0000

Main reason to me is that its layers on layer on top of the base LLM calls with not so much to show for it. Also a lot of native features (like for examples geminis native structured responses) aren't well supported.

New comment by lysecret in "Ask HN: Why there are no actual studies that show AI is more productive?"

lysecret — Sun, 08 Mar 2026 10:22:20 +0000

Because we are incapable of measuring developer productivity.

New comment by lysecret in "Claws are now a new layer on top of LLM agents"

lysecret — Sat, 21 Feb 2026 11:35:32 +0000

Im honestly not that much worried there are some obvious problems (exfiltrate data labeled as sensitive, take actions that are costly, delete/change sensitive resources) if you have a properly compliant infrastructure all these actions need confirmations logging etc. for humans this seemed more like a neusance but now it seems essential. And all these systems are actually much much easier to setup.

New comment by lysecret in "Gemini 3.1 Pro"

lysecret — Thu, 19 Feb 2026 18:37:55 +0000

Please I need 3 in ga…

New comment by lysecret in "Show HN: LocalGPT – A local-first AI assistant in Rust with persistent memory"

lysecret — Sun, 08 Feb 2026 09:40:31 +0000

Local really has a strange meaning when most of what these things do is interact with the internet in an unrestricted way

New comment by lysecret in "Show HN: Moltbook – A social network for moltbots (clawdbots) to hang out"

lysecret — Sat, 31 Jan 2026 16:48:48 +0000

It’s just fake accounts being created the bot network has a bot problem haha

New comment by lysecret in "Show HN: Moltbook – A social network for moltbots (clawdbots) to hang out"

lysecret — Fri, 30 Jan 2026 17:05:57 +0000

There is son much personal info in here it’s wild.

New comment by lysecret in "Show HN: OfferGridAI – side-by-side comparison of real estate offers from PDFs"

lysecret — Fri, 02 Jan 2026 18:41:51 +0000

Hmm so that post also links back to segmentation done by structured outputs? (Though here not even enforcing the structure)

https://ai.google.dev/gemini-api/docs/image-understanding#se...

New comment by lysecret in "Show HN: OfferGridAI – side-by-side comparison of real estate offers from PDFs"

lysecret — Fri, 02 Jan 2026 14:23:31 +0000

Oh I didn’t know about the visual bounding boxes this is super cool!

Quick question are you talking about this feature?

https://docs.cloud.google.com/vertex-ai/generative-ai/docs/b...

Because it’s just using structured response so it should be doable with Gemini 3 ? (We are using Gemini 3 for some docs processing and its visual understanding is just incredible)

New comment by lysecret in "LLM Year in Review"

lysecret — Sat, 20 Dec 2025 11:26:42 +0000

It’s funny how every podcaster/public ai figure is so certain text as a Ui will go away and it’s not going anywhere.

New comment by lysecret in "Mistral OCR 3"

lysecret — Sat, 20 Dec 2025 09:10:48 +0000

3 flash is also insanely good even slightly outperforms 3 pro for me.

New comment by lysecret in "Announcing the Beta release of ty"

lysecret — Wed, 17 Dec 2025 08:28:56 +0000

Super excited about this generally ok satisfied with pyright but so I was with conda before uv or black before ruff.

New comment by lysecret in "I tried Gleam for Advent of Code"

lysecret — Sun, 14 Dec 2025 12:53:02 +0000

When I was debugging through f# code they definitely had that.

New comment by lysecret in "Getting a Gemini API key is an exercise in frustration"

lysecret — Thu, 11 Dec 2025 11:09:09 +0000

I see this, a hot take form my side as someone who is bought in to GCP i quite like being able to put everything on the same billing account / handle it easily through service accounts.

New comment by lysecret in "Gemini 3 Pro: the frontier of vision AI"

lysecret — Sat, 06 Dec 2025 10:44:24 +0000

2.5 pro is already excellent at this.