jorgeleo

I have so many mixed feelings about this...

On one side, as an engineer, Did you really expected to delegate thinking to a Large Language Model? On the other side, How human of the author to believe the illusion that fluency means knowledge, can't fault him for that.

On one side, so you discover that the development of your personal skills cannot be distilled to a set of instructions to an LLM? How much more of you are you willing to capitulate? On the other hand, I think is brave to write the article confessing his growth.

There is a generation that has been trained using the current teaching methods where education in it self is a product as opposed the gym where the brain develop muscles. The muscles end up been develop at work. This is not about "seniority", this is about the character to develop yourself, to welcome the challenge. Which I guess good for him, he has it.

Reading these kind of things for me is weird.

(btw. I don't like the term AI because there is no I in AI, but that is a different talk, or maybe that self awareness is what makes it so confusing)

My experience trying to get better coding out of LLMs, and how I walked towards improvement.

This is the very reason why I avoid to use native harness. They are optimized for the economic benefit of the provider, not for mine. Love Pi because it does a good job managing the context, open code meh, claude code nope.

Same thing for me. on an M5 max, Qwen 3.6 35b gives me between 150 and 200 tps using splash as inference engine.

More than enough for guided code sessions, at 100% privacy. And i can use obliverated models if i am trying to harden my own app, something i cannot do with cloud providers.

An itch that I have been trying to scratch for a while: I need a scripting language that I can integrate into a web app, and I can debug from the web app. It doesn't need to be complex or with curly braces, actually the simple the better.

Three days with AI and FluidScript is born. The language is minimalistic, the keywords are simple, and you can stop execution at any line and you get the whole context back, later you can continue from that point passing the context, and it continues as if it hasn't stopped. Full host integration and a Monaco editor example that shows how to do break points.

Dreamed it for years, now, once I have a clear concept and design, AI type it on 3 days.

Yup!. Over 30 years experience here. Landed on the same place with the added feeling of "We were always intended to develop the judgement of what makes sense and maintainable, where is the surprise here?"

People want the "Fluent Text Generator" to think for them because its marketing calls it AI. At the end is just a tensor collection, probabilistically backtracked adjusted, text generator. It does produces fluency in the answer, but the fact that it sounds correct, to the point of passing a compiler approval, has nothing to do with been well thought out.

curios minds... why to do that port?

i can sell you my MBP, but no refunds if it doesn't work

hahaha

and redo the thermal paste

I have personally replaced the battery a couple of times already... but a good clean up is a good idea

i decided to do an experiment and try to run an LLM in my old 2013 MBP. i7, 16 gb mem, 1 tb hd.

Installed Linux mint Xfce Edition for lightness, installed ollama, start to test different models. Gemma4 e4b runs perfectly fine, exposed it to the network, connected to it with my current notebook and use vs code codex to start to run inference.

For about 30 minutes of bliss, this setup work at a reasonable speed... then the MBP shut it self down. It was so hot that it trigger the safety mechanism, the fans sounded like the laptop was about to take off.

I though on leaving it on inside the fridge, but then the WIFI wouldn't reach.

On the other hand, my wife saw all this and offer to buy me an M5... the experiment didn't work as intended, but it did work.

I was all for trying it until I saw this in the License Agreement:

"4.1. Zed's Use of Customer Data Customer hereby grants Zed a non-exclusive, worldwide, royalty-free, fully paid-up, non-sublicensable (except to service providers and Customer’s designees), non-transferable (except as set forth in Section 15.1) right to use, copy, store, disclose, transmit, transfer, display, modify, create derivative works from, collect, access, store, host, or otherwise process (“Process”) any materials that Customer inputs into or otherwise makes available to the Service (including prompts and other written content) (collectively, “Customer Data”) solely: (a) to perform its obligations set forth in the Terms, including its Support obligations as applicable; (b) to derive and generate Telemetry (see Section 4.4); and (c) as necessary to comply with applicable Laws. Except as required by applicable Laws, Zed will not provide Customer Data to any person or entity other than Customer’s designees (including pursuant to Section 7) or service providers."

Sorry, no I don't agree to make my source code and the product I am working to give you "non-exclusive, worldwide, royalty-free, fully paid-up, non-sublicensable (except to service providers and Customer’s designees), non-transferable (except as set forth in Section 15.1) right to use, copy, store, disclose, transmit, transfer, display, modify, create derivative works from, collect, access, store, host, or otherwise process (“Process”) any materials that Customer inputs into or otherwise makes available to the Service (including prompts and other written content)"