itmitica

7 ms·

61.
by itmitica 8mo ago
The question remains: no quantization of the confidence level Mesa is adding or subtracting. Mesa enforced rules may well be ignored, is this true? Mesa is yet another exercise in passing the responsibility to a magical system. Tests and hu
62.
by itmitica 8mo ago
How does it stack up? https://www.businessinsider.com/ai-agents-study-company-run-... https://kpmg.com/nl/en/home/media/press-releases/2025/11/ai-...
63.
by itmitica 8mo ago
Why make others misfortune a platform for ego expression? Why not doing things elegant, quiet, keep it in-house? Because misery of others drives stock prices up! It's a sacrifice he's willing to make.
64.
by itmitica 8mo ago
I'm sold! But, what about...
65.
by itmitica 8mo ago
Again, you are building an audit agent. You just use some theater around it. For the purpose, the best audit agent is a completely different agent, not a different persona of the same agent.
66.
by itmitica 8mo ago
So, you're saying, build tests? TDD?
67.
by itmitica 8mo ago
I did the exact same thing! Uncanny. I agree with models being better at different tasks: gemini-cli is superficial, codex is stubborn as a mule and dependable, claude-cli just wants to get something working and done. qwen-cli, Qwen, in gen
68.
by itmitica 8mo ago
... or, maybe, simply use another agent to audit it? Agent teams sound better than literature-induced confidence. Are you anthropomorphizing when you should just automate the review?
69.
by itmitica 8mo ago
I appreciate your work. But, again, from a productivity point of view, and from a correctness of approach point of view, I have learned this: 1. Avoid overengineering against/at all costs. 2. Doing the project is doing the project, any
70.
by itmitica 8mo ago
I like your content very much, let me point this out first. I'm not sure all aspects are covered in the approach. For instance, controlling the agents takes a big chunk of the interest. The agentic system architecture is also big in vi
71.
by itmitica 8mo ago
I too was convinced at one point the spec is the program. That it doesn't matter the implementation stack. But, after wasting too much time in the meta, with nothing really to show for, I returned to controlling the programming process
72.
by itmitica 8mo ago
Interesting project. Interesting readings in the project, such as https://github.com/desplega-ai/advanced-context-engineering-... . I'm not sure why, but I keep trying to reject this, subconsciously. Like, there is
73.
Kyoto University develops AI monk robot equipped with Buddhist scriptures(english.kyodonews.net)
2 points by itmitica 8mo ago|0 comments
74.
by itmitica 8mo ago
Not really. You just found another way to tell Claude "I want a game, I am uncertain, let's randomly seed the start of the process.". But you did it circus-style :)
75.
by itmitica 8mo ago
Code Mode is a workaround for poor tool packaging. Instead of improving the MCP ecosystem’s capability discovery model, they reduced context pressure by collapsing the tool ecosystem into a programmable backend. It’s a vendor-specific, host
76.
by itmitica 8mo ago
“We won’t fix the simple, visible server-side problem, so we’ll distribute a harder version of it into every client.”
77.
by itmitica 8mo ago
Flawed. GPT 4.1 gets it right. GPT 4.1 mini answers wrongly. It's about quantization, not about model. The companies clearly cut corners on some inferences, they are quietly using lesser models than advertised or listed in plain sight.
78.
by itmitica 8mo ago
The normal reaction is, there you are, I managed to do this using cli. The gratuitous ragebait way is, there you are, I managed to do this using cli, hence MCP is dead. But why?
79.
by itmitica 8mo ago
În real world, worst performers get thrown out of the loop if they are flagged and fail to improve. Assumption here seems to be agents have no such issues. Am I missing something? In the real world, over engineering is the play to work-pret
80.
by itmitica 8mo ago
All models are trapped in memory in an ephemeral space of time. Agents used to interact with the models make it seem otherwise.
81.
by itmitica 9mo ago
Share and you can't control what everyone else does with the data. Hence, untether is a non concept. In the sense that that data has now lost it's uniqueness. You can untether all you want. Copies of it will be subjected to any ki
82.
by itmitica 9mo ago
To share is to lose control. You can't undo, even once shared, it can't be undone. You can't retract a published novel. You can't retract a broadcast music or show. What makes you think you can do it over internet?
83.
by itmitica 9mo ago
JSON Schema? https://github.com/rjsf-team/react-jsonschema-form
84.
by itmitica 9mo ago
Is this a bad bait or is it a bad post? I can't decide.
85.
by itmitica 9mo ago
hmmmm....
86.
by itmitica 9mo ago
Seems like the natural thing to do. Also did that already, offered all but the code and interposed bring-your-own-AI in the process, and seems like a natural convergence.
87.
by itmitica 9mo ago
Yeah, but still not org mode.
88.
by itmitica 9mo ago
Question: why is ` https://orgmode.org/ ` in html and not in ... org mode?
89.
by itmitica 9mo ago
Syntax highlighting gains, regardless of color scheme, are not negligible. There are grayscale color schemes, for example, that remove the violent nature of some color themes.