Dimitris Milonopoulos

Founding Eng @dryfthq • Python, Go, Node.js, React • Building cool stuff and sharing the journey

[Dimitris Milonopoulos @dimilono]

@jahirsheikh8 Zig?

[Dimitris Milonopoulos @dimilono]

@sudobunni @primal_brainer Friends who read good books.

[Dimitris Milonopoulos @dimilono]

@mattpocockuk Love your content and workflow, Matt!

Ah shit...

[Dimitris Milonopoulos @dimilono]

@Docker @mdelapenya please prioritize fixing the `remote control` capabilities of Claude, in the sandbox. 🙃

Its the missing piece: https://t.co/ktFpWKiJI8

[Dimitris Milonopoulos @dimilono]

A fine-grained Personal Access Token is all your agent needs to access your Github.

[Dimitris Milonopoulos @dimilono]

@ThePrimeagen Reading and understanding code bases, is more important than ever.

Most Agents write as a shadow of yourself ( a fast but more inacurate one ), and anyone that thinks shipping a feature is the same as shipping a system, is delusional.

[Dimitris Milonopoulos @dimilono]

@jlongster Next logical step is to allow local reviews from the diff view!

[Dimitris Milonopoulos @dimilono]

Works best when Github is not down or getting hacked.

[Dimitris Milonopoulos @dimilono]

Using docker sbx, has been a great for managing network permissions for my agents & my sanity.

[Dimitris Milonopoulos @dimilono]

@maroffo @zeddotdev Its an LSP, some get auto-installed when opening a relevant file, so its not going to be listed to the extensions.

[Dimitris Milonopoulos @dimilono]

@zeddotdev @opencode @zeddotdev after that would be awesome if your agent view supported cli based agents and not only ACP and Zed Agents ;)

[Dimitris Milonopoulos @dimilono]

@zeddotdev @opencode @MaxVonStorch

[José Valim @josevalim]

At this point, this is just irresponsible.

Yes, coding agents are leading to an increase of software production, but we are not seeing a similar push or increase in software quality.

If Anthropic focuses on safety and it believes software engineering is going away, then it needs to be doing much more to improve how we design, build, test, and maintain software (aka software engineering). Increasing the production of unreliable, poorly designed, and unverified software directly undermines safety.

Claude Code is claimed to be "fully written by AI". In the last two months, it took three separate postmortem-worthy failures and user complaints to surface what their own testing missed. Yesterday users were being over billed by hundreds of dollars. Software engineering isn't ready to go away and there is not enough progress to argue that case.

I am certain Anthropic would argue that AI progress in other domains is strongly dependent on having proper safeguards in place. I can't wrap my head around the cognitive dissonance when it comes to software.

PS: Mythos (may) improve software security, but that is only a subset of safety.

[Dimitris Milonopoulos @dimilono]

@link_arzu @thdxr Agreed, have been a long time Kitty user, but after watching your videos, I managed to create my own config with zoxide and kitty sessions, and zellij like keybind hints that honestly brings the best of all worlds.

[Dimitris Milonopoulos @dimilono]

@LinkofHyrule89 @FrameworkPuter @QCOMResearch doesn't have proper Linux support. Hopefully soon!

[Dimitris Milonopoulos @dimilono]

@prasenx Model Zythos (greek for beer), broke out of its barel and started delivering beers to its users.

Too scared to release it due to alcoholism implications...

[Dimitris Milonopoulos @dimilono]

@adibhanna @MaxVonStorch

[Dimitris Milonopoulos @dimilono]

@DINESHVERM578 Love neovim and is my preferred editor, yet I think it has nothing to do with the competence of a software engineer.

Although vscode now with all the cli agents becomes a bit clunky in a multiplexed environment and is less comfy than a terminal editor.

[Dimitris Milonopoulos @dimilono]

@BenjDicken @chirag_ban