Viewing profile — vishnukool
vishnukool
HN member- Joined
- Sat, Feb 16, 2019, 6:21 PM UTC
- HN karma
- 101
- Public activity
- 80 items
- HN profile
- View on Hacker News ↗
About vishnukool
Recent public activity
-
comment
Comment #48993163
I built basically this on top of my telegram setup last year. The thing that killed it was group chats, per-message LLM calls on a busy group eat your api budget alive, so I ended …
-
comment
Comment #48988650
Such a rage bait title, but sure I'll take the bait. It is easy to phrase anything as slaves. Are refrigerators and microwave ovens our slaves? But if you push the tech fare enough…
- story
- story
-
comment
Comment #48232433
The problem is that LLMs do not have a conceptual grounding in actual time. They estimate based on statistical correlation found in their training data which is filled with standar…
-
comment
Comment #48181744
Good one. One thing that's becoming clear is that agent security is less about jail break prompts and more about permission boundries, hidden context flow and un intended tol behav…
-
story
Ask HN: Are all the YC founders lying offlate?
Ask HN: Are all the YC founders lying offlate ? I've been seeing more and more examples of things like these where apparently YC founders are just downright fraudulent or lying abo…
- story
-
comment
Comment #48118852
System design definitely carries more weight now but Leetcode has not totally vanished. It's just evolved.
-
comment
Comment #48085450
Respect. It's a good day to remember the impact they had even if they are no longer around to pick up the phone. Btw calling her now.
-
comment
Comment #48060161
This is the weird part of AI hallucinations and people start treating it like a verified fact after it gets repeated.
- story
-
comment
Comment #46995871
What have you seen to be the bottleneck ? From the usage patterns I've seen for 90% of the use cases it seems to be fine, the one exception being Browser based use cases which chug…
-
comment
Comment #46974917
Very true and actually totally spot on! It's the damn chromium instances that chugs memory like an Irish man chugs beer. I experimented launching containers with it and quickly rea…
-
comment
Comment #46974895
Yes exactly! Although we've launched a Claude Skill to help setup OpenClaw! So Agent 37 is literally the easiest Openclaw setup at this point :)
-
comment
Comment #46974882
Absolutley! We do have some metrics like, when a user interacts with a skill or not, and so that's a signal
-
comment
Comment #46974876
Yes, once we have a large enough numbers to justify that we'll factor that into the search results quality
-
comment
Comment #46974870
Ah that's a good point, I think I could introduce some kind of time decay by also factoring in last updated repo time into the search priority.
-
comment
Comment #46974858
Great point, we de-duplicate the search results by Repository Id. that largely filters out content from the same person , and note that search prioritizes github stars, so low qual…
-
comment
Comment #46974841
Thank you! We have limits around usage. Also as a startup we have a lot of Gemini credits from Google, so that helps in offering it. Not to mention the gemini 3 flash that it's run…
-
comment
Comment #46969908
Yup we actually backup data in a separate partition per container. So it survives restarts and even accidental deletions.
-
comment
Comment #46969903
Thanks! At the moment no, but down the line I imagine we can expose it if this is useful
-
comment
Comment #46969901
We do back up data, although it's manual. The docker instance home directory and homebrew are backed up in a separate partition. So it survices restarts and container updates.
-
comment
Comment #46969892
SO far pretty good! Turns out these things are mostly idle
-
comment
Comment #46969889
We only allow ttyd not ssh, so users only have access to their individual containers and isolated from the host. But inside the container they can do whatever they'd want. There ar…