"Andy" Antti Törrönen

Move fast and fix things. Fix the world. 🌎 🌍🌏Founder at Kwork Innovations. Building applied AI systems & quantitative investing.

["Andy" Antti Törrönen @torronen]

@thsottiaux comfort in using without external mouse, especially scrolling, finding bars, dark mode. Not terrible, but annoying.

Annoying: threads disappear...

Not annoying but would be nice: subfolders + folder hierarchy.

Power user move: additional description and nicer view of threads.

["Andy" Antti Törrönen @torronen]

@thsottiaux Is it open for anyone to build such tools?

["Andy" Antti Törrönen @torronen]

@garrytan okay now I know what you need the 10+k LLM credits for, wish you luck! About one every hour or so...???!

["Andy" Antti Törrönen @torronen]

@KadiriZaid @thsottiaux i had the same but the reason was i was routing the traffic... working from home connected to office and the traffic was routed through office firewall

["Andy" Antti Törrönen @torronen]

@thsottiaux any chance the macos desktop app could also connect to codex on running on another machine? I hate my battery runs out because it decides to run all tests in the project a few times over.

["Andy" Antti Törrönen @torronen]

@jarimh1984 Did you have a chance to review how is the privacy policy? Can we try it on codebases that we should not be trained on?

["Andy" Antti Törrönen @torronen]

@Dystrophical @rbkayz @MercuriusFilius so it could be better or worse... an unknown risk. That is a good additional argument not to bid.

["Andy" Antti Törrönen @torronen]

@s_streichsbier Just constant: "Error running remote compact task: stream disconnected before completion: error sending request for url (https://t.co/oUjoiVbiGU)"...

["Andy" Antti Törrönen @torronen]

@thsottiaux PRIVACY. Privacy shield icon: am i being trained in style of copilot. Temporary projects. Delete logs. Easier to understand where log are and for how long.

["Andy" Antti Törrönen @torronen]

@thsottiaux Translations need polish. They can do more harm than good if not careful.
e.g.: Finnish "sopimus" = contract,
should be "suunnitelma" = plan

Took me a while to figure out which contract can be enabled.

["Andy" Antti Törrönen @torronen]

@thsottiaux it sends all queued messages even if quota is empty. Could it please check the quota before trying to send a message so I can just make it continue when quota renews

["Andy" Antti Törrönen @torronen]

@thsottiaux scroll with touchpad and/or arrow keys could be better on macbooks. Maybe have someone use it with just a keyboard for a full day and plan how to make it perfect. latest MacOS, dark mode.

Main issues: scroll chat window, scroll project list, find scroll bars, scroll long diffs..

["Andy" Antti Törrönen @torronen]

@GaelBreton ran out of credits, sorry, will complain soon again after reset

["Andy" Antti Törrönen @torronen]

@melvynx True. I reserve max 20x for specialist tasks (same as gemini cli) while codex is 90%

["Andy" Antti Törrönen @torronen]

@MichaelMatcha @_tenZdhon_ True, LinkedIn is now obsessed with Mythos and other A.I. releases. Only thing that exceeds light of speed is speed of people becoming A.I. consultants.

TBH, I get it. A.I. needs people from all fields and, besides, times are hard.

Anyway, we just like to make fun of LinkedIn

["Andy" Antti Törrönen @torronen]

@kr0der what is wrong with that? Some developers prefer it that way. I am not sure I see anything wrong with layers of abstractions and fallbacks in case a method fails. It is a bit lazy but at least one of the fallbacks works out-of-the-llm.

["Andy" Antti Törrönen @torronen]

@thekitze 5.4. for work that matters, codex spark for review, quality passes and specific tasks where I just want it to get over with. Is there a reason for non-spark 5.3?

["Andy" Antti Törrönen @torronen]

@mean_field_zane Realizing that is the lesson, I guess

["Andy" Antti Törrönen @torronen]

@sughanthans1 @joburgai @TheAhmadOsman true, but earlier 'they' couldn't use our data, such as emails, in mass. They could find trade secrets but now cloud providers, theoretically, could just take our data an train on it. Theoretically, of course, because it is not like anyons is obsessed with winning the AI wars.

["Andy" Antti Törrönen @torronen]

you take your dirty claws off my tokens!
🦞🦞🦞🦞🦞🦞🦞🦞🦞🦞