ramoz

11 ms·

181.
by ramoz 7mo ago
My nervous system is damaged, unfortunately, therefore it's still here! I appreciate the tips though!
182.
by ramoz 7mo ago
For me, it's a very distinct ringing. Like in the movies when there's some explosion/shock scene and there's a very persistent ding that happens after the explosion or whatever - it's what I hear constantly.
183.
by ramoz 7mo ago
> Claude Code style plan mode to be a bit restrictive Hey thats why i built plannotator: https://github.com/backnotprop/plannotator I like staying within Claude Code for orchestrating its plan mode, but I needed a b
184.
by ramoz 7mo ago
https://x.com/benjitaylor/status/2027902450049708385?s=46
185.
by ramoz 7mo ago
Don’t let it get to you like this.
186.
by ramoz 7mo ago
Same experience here.
187.
by ramoz 7mo ago
I thing the real impact behind the scenes here is Bash(). Filesystem relevance is a bit coincidental to placing an agent on an operating system and giving it full capability over it.
188.
by ramoz 7mo ago
If you don’t use VSC, don’t enjoy emoji slop, and also want to see sub agent log chains: https://github.com/eqtylab/agent-console Someone else’s more polished Mac app: https://readout.org
189.
by ramoz 7mo ago
curl -sL https://github.com/openai/gpt-2 | grep -iE 'Star |stars' This still shows star data, so I feel like this is probably a bug.
190.
GitHub appears to be hiding repo stars for signed-out users
3 points by ramoz 7mo ago|1 comments
191.
by ramoz 7mo ago
> the terminal is the most powerful way to interface with the operating system. There's something about the UX here, how we're interfacing in an abstract way directly with the computer, now with agents. I hope we can get some i
192.
Faulty reward functions in the wild (Jack Clark, Dario Amodei, 2016)(openai.com)
4 points by ramoz 7mo ago|0 comments
193.
by ramoz 7mo ago
It was typically isolated in procurement, only buying data from "outside of the border" - data was packaged by geo regions typically
194.
by ramoz 7mo ago
From my experience this data is not cheap from an average consumer perspective.
195.
by ramoz 7mo ago
I worked closely to some of this. There were strict policies in place to never monitor US Citizens. That said i was focused in more kinetic warfare domains and not sure what would've extended past the borders by local law enforcements
196.
Low data gravity for fast retrieval on K8s (2021)(ra-mos.medium.com)
2 points by ramoz 7mo ago|0 comments
197.
by ramoz 7mo ago
nice thoughts
198.
by ramoz 7mo ago
> consensus best practice I'm not sure I agree with this. I don't think there needs to be a whole spec & documentation process before plan mode. There is alternative thought leadership that the waterfall approach for buildi
199.
Helpme: Minimal tmux wrapper for context-aware agent debugging(github.com)
3 points by ramoz 7mo ago|0 comments
200.
by ramoz 7mo ago
> what testing should be built Yea, a big part of my planning has included what verification steps will be necessary along the way or at the end. No plan gets executed without that and I often ask for specific focus on this aspect in pla
201.
by ramoz 7mo ago
My workflow has become very plan-intensive... including planning of verification+test steps at the end.
202.
by ramoz 7mo ago
Beautiful landing page
203.
by ramoz 7mo ago
Quite a bit. - Research - Scan the web - Text friends - Side projects - Take walks outside etc
204.
by ramoz 7mo ago
I did a sort of bell curve with this type of workflow over summer. - Base Claude Code (released) - Extensive, self-orchestrated, local specs & documentation; ie waterfall for many features/longer term project goals (summer) - Base
205.
by ramoz 7mo ago
OpenClaw is not going away anytime soon. And I don’t think it can be platformed behind web UIs. OpenAI owns an OS with this one. I avoided the hype at first; however, it has become extremely efficient for emails and notes, and I can see how
206.
by ramoz 7mo ago
Not sure why you're getting down voted. This is totally reasonable.
207.
by ramoz 7mo ago
We think so as well with emphasis on "why" for commits (i.e. intent provenance of all decisions). https://github.com/eqtylab/y just a prototype, built at codex hackathon The barrier for entry is just inclu
208.
by ramoz 7mo ago
why was it flagged?
209.
by ramoz 7mo ago
Yea I spend a lot of time in plan mode, annotating/iterating on plans before I feel good to hit go.
210.
by ramoz 8mo ago
Im terrible with email, so its be genuinely helpful for me there. Excited to explore more use as time permits. Very optimistic based on email experience. My next use case is personal notes system.

More ›