Joe Muoio @JoeMoCode

Full stack freelance software developer and consultant building AI integrations and Cloud services. I like games and food. blog.joemocode.com Seattle, Washington Joined June 2019

Tweets

776
Followers

665
Following

617
Likes

2K

Joe Muoio @JoeMoCode

6 days ago

@mattpocockuk Yeah I am stuck on this too. The way I think of it... Check in and maintain only what needs to be evergreen. Then for feature branches, I check in semi-ephemeral artifacts which includes research and plans. But need to archive or remove these when finished. External KB good, too

0 0 0 34 0

View Details

Joe Muoio @JoeMoCode

3 weeks ago

@plainionist It has always been a context problem. And dare I say, always will (with LLMs). You can never rely on the latent info in an LLM and it must always be grounded. Just from the basics of how it works and realities of changing context and training pipelines.

0 0 1 50 0

View Details

Joe Muoio @JoeMoCode

3 weeks ago

Got an ASUS Ascent GX10. What are your top 3 local models to run and why? So far, playing with qwen 30B instruct coder model and pretty amazed with how fast it is and generally quite good with tool calling.

0 0 0 76 0

View Details

Joe Muoio @JoeMoCode

a month ago

Maybe 4.8 is "better" at coding, but I will just start trying it for planning for now. I'd much rather use a cheap model for implementation than an expensive one. Big context window and better natural language processing is useful when researching and planning, not coding.

1 0 0 37 0

View Details

Joe Muoio @JoeMoCode

a month ago

Opus 4.8 is out, but I am WAY more hyped about Composer 2.5 for coding. Opus class has been good for planning... but it's somehow worse at coding than composer. By the time I code with AI, the task is well thought through and I just need it done.

1 0 1 66 0

View Details

Joe Muoio @JoeMoCode

a month ago

@plainionist yeah totally there is a place for rules. I see them primarily as a way to direct agent to the place it needs to find that info rather than capturing it in a likely duplicate way. I don't give agents roles like that so I don't have opinions on that, specifically.

0 0 1 22 0

View Details

Joe Muoio @JoeMoCode

a month ago

@sickdotdev 100% Cursor Vertically integrated harness has most potential, but also most risk when the company is built to sell compute VSCode (with devcontainers) + Cursor CLI for in the loop dev. Cursor Cloud agents for parallelized sessions. Cursor IDE is not my cup of tea

0 0 0 168 0

View Details

Joe Muoio @JoeMoCode

a month ago

github.com/mattpocock/ski… Matt's skills are great. At least the few I have used, so far. Lots of good stuff in here I have not even discovered yet, I am sure.

Matt Pocock @mattpocockuk

a month ago

Skills should be: - Concise - Responsible for one thing, not multi-step - Composable - Progressively disclosed - Harness-agnostic What else? Or - what did I get wrong?

190 82 2K 134K 1K

0 0 0 65 0

View Details

Joe Muoio @JoeMoCode

a month ago

@plainionist not everything is so simple, but there are some things that are :) Rules are always gonna be "okay-ish" and never going to SOLVE a problem. Only mitigate how often it happens.

0 0 1 11 0

View Details

Joe Muoio @JoeMoCode

a month ago

@plainionist That's insulting to the goldfish who do actually have memory! Shift away from rules to determinism. Example, don't have rules about code style, instead have a hook that runs your linter and hook that up to run when a file is changed by your AI harness. No need for style rules

2 0 0 45 0

View Details

Joe Muoio @JoeMoCode

a month ago

@DanielMiessler @MattTrifilo And this is a core assumption of the argument. That we have "intelligence". I like that you defined it here and agree models give us this to an extent. Employees also give accountability. How okay is it if the workflow is wrong? who has accountability? Bigger problem, IMO

0 0 0 37 0

View Details

Joe Muoio @JoeMoCode

a month ago

@thdxr I been thinking this as well... but, don't forget about your own skill growth in that time! Turns out things are hard. Software is hard, and we are here and like it because it is hard. Always something new to learn and apply.

0 0 1 365 0

View Details

Joe Muoio @JoeMoCode

a month ago

Only if the practitioner wants better software and knows what that means and what it takes. Lots of examples to the contrary -- more bad software getting made than ever before. More great software being made as well. ...just more software getting made. :)

Seb @plainionist

a month ago

Before AI, software development felt like: 10% design - 80% coding - 10% testing With AI, it became more like: 40% design - 20% prompting - 40% testing More thinking. More verification. So ... better software? 🤔

39 1 56 4K 5

0 0 0 21 0

View Details

Joe Muoio @JoeMoCode

a month ago

@plainionist huh, that's a very interesting tool... 🤔🤔🤔

0 0 1 5 0

View Details

Joe Muoio @JoeMoCode

a month ago

@plainionist DSL needs to REALLY show value for it to be adopted and used to the point that we then fine tune models on it at scale. Without that, you are locking yourself in and/or the model will have minimal knowledge of it, or will have to bloat context to share knowledge of DSL.

1 0 1 22 0

View Details

Joe Muoio @JoeMoCode

a month ago

@plainionist For runtime use cases with claude for instance, I use XML in order to better separate parts. Recommended in claude docs, too. I don't use it for general plans/notes/research and still opt for md. Kinda since its the default format and I haven't felt the need yet to change.

1 0 1 27 0

View Details

Joe Muoio @JoeMoCode

2 months ago

Clarity is good. Changing how your tool works in a way that is not technically sound and breaks the work folks who love building on your tech and pushing subscriptions is exactly the reason why I left Anthropic behind months back. Big miss from Anthropic in a year of misses.

ClaudeDevs @ClaudeDevs

2 months ago

Starting June 15, paid Claude plans can claim a dedicated monthly credit for programmatic usage. The credit covers usage of: - Claude Agent SDK - claude -p - Claude Code GitHub Actions - Third-party apps built on the Agent SDK

1K 1K 13K 10.9M 6K

0 0 0 58 0

View Details

Joe Muoio @JoeMoCode

3 months ago

@nbaschez Same problem we always had as devs with analysis paralysis. The point of the plan and doc is to think through the problem. So you are done when you have done that and have a sufficiently detailed written plan you approve of.