oulu2006

Apple playing 3d chess here -- they can just absolutely kill it in making the HW available for everyone to run models at home, and will do a better job than Nvidia especially when M7 comes out which is specifically designed for AI workloads.

Thanks OP

So everyone has their preferred way of doing it.

1. I go direct to source, i.e. DS platform, I find it cheaper than paying the openrouter tax -- I also switch it up a bit

2. I built a local LLM router, that I update with new profiles that have my preferred provider of the week (lowest token costs/speed) with fallbacks, like mimo --> DS4 etc.. if there is overloading,

3. I use 3 diff harnesses, CC/Codex + Opencode -- they all talk to each other through a custom rig system that routes messages between llms using a Rust backed structured JSON system

Not saying this is the best, it's just what I like and works for me^.

I can flow quite naturally between Opus/Astra/K3/GLM/MiMo/DS/etc.. this way and often do...more so these days with subs no longer great as they used to be.

yeah same that's my preferred flow now -- sill use Claude and OpenAI a bit, though recent nerfs to subs has really made using codex much harder.

DS 4.1 flash is my main powerhouse and Opus/Astra my auditors (when they're not out of tokens) otherwise K3 or DS4 pro

See my later post in this thread, I go into more detail about how I use AI in the restorative process.

In addition, I literally had AI build a complete list of parts to order from Amazon down to the caps for each system I wanted to rebuild -- it was close to 20 different items, I would have probably given up if I had to try and assemble this list from different YouTube videos.

No, not everyone, there's no easily accessible video showing how to setup aquarium lights to slowly de-yellow ABS plastic -- at least I didn't find one; AI helped me find an obscure reference to it and I then read a few message boards where it was mentioned.

I also use AI to take in-progress pictures as I desolder to help me check for traces that need to be repaired or help identifying specific chips. I probably could try and find a video where the same chip is featured and someone explains it, and/or retrieve the schematic for the specific logic board, but that's very painful and does slow the process. Think of AI, in this specific case, as enabling skill development for me in a field I wouldn't have necessarily have gotten into, because of being short on time and AI helps me consolidate that information quickly.

I'm learning by doing things.

For example: AI has helped me get into restoring retro tech, specifically resoldering leaky caps on retro Macintosh logic boards. Before AI, I didn't know how to use a multimeter (I knew theoretically how it worked), I didn't know how to use flux, solder wick, heat gun. I also didn't understand how bromine radicals yellowed plastic and how to reverse it by using blue light similar to what they use for indoor aquariums.

So AI unlocked doing for me.

or https://sydneygigguide.net/dl/893a15 in Sydney.

This is an interesting comment -- your life is precious brother, you might have something in store down the road :)

Precisely, there are about 100 of these, and everyone makes a new one every week.

He's correct internationally, compared to real socialists in countries like Australia/NZ, US dems are definitely mildly center right.

This is more accurate, I've written enough code in my life to never really want to do it again ....but I still love creating (code was merely the way to do it) so LLMs help with my underlying passion.

I don't find the same, like you, principle/CTO engineer, there's a world of difference between simplistic prompt/vibe coding and building a properly architected/performant/maintainable system with agentic coding.

That's great and I'm the same, 40s multiple founder and I was ready to hang it up after my last exit -- had 0 passion to code anymore and now I'm back and LLMs are reigniting my passion to create again.

It's Qwen3.5 now, you're a bit behind the times.

Same, I have a bunch of skills defined ith proper YAML headers and semantic triggers installed, I make a point of listing not too many but making it quite specific.

Even with that, I have to be very specific in triggering a skill and it's hit or miss if it picks up on the skill -- usually I have to say there is a skill with this go and use it.

That's really interesting, I love the idea of being able to use columnar support directly within postgresql.

I was thinking of using Citus for this, but possibly using duckdb is a better way to do. Citus comes with a lot more out of the box but duckdb could be a good stepping stone.

> production, or reduced execution speed by 50%

Hopefully that's "reduced execution time" :)

I like the remarkable paper pro -- been using it for 8 months consistently.

The rest of the world just shakes its head at the US healthcare system.

Powerful forces keep it as it is -- and people seem to buy the stupid communism/socialist crap that they pedal and soak it up hook line and sinker.

Gave up trying to understand why.

Never left bittorrent.

I really wanted to embrace streaming services, prime really killed me recently, with introducing ads into a membership that I already pay for! and 90% of movies on prime I have to pay an extra 20 bucks to view ...why am i paying for membership?

What an incredible read, the MS-DOS revelation is wow

right people mostly, I partnered with someone different for the 3rd startup; from my experience with the first 2 i realized the sort of person I needed as a co-founder.

Someone to compliment my deficiencies, domain expertise and with the same focus and mindset to succeed -- most startups fail because of founder issues.

that was interesting, thanks for the reference

3 products

1st one didn't sell

2nd one sold for 000s

3rd one sold for 000000s

4th one ?

so really went the other way, quality over quantity

not sure if that's a bad thing when it comes for FE frameworks - not reinventing it every 5-7 years i think is a good thing.

I just do it with multiple terminals + checkouts + branches -- this seems to just get in the way.

But I'm an old school vim/emacs keyboard only hacker, so terminal all the way is super efficient for me

I'm already building whole projects now with AI -- not a single line of code by myself.

I love it, I've coded for 25 years and I don't need to write another line ever again, I just want to build cool things as fast as possible.

not just employees, founders as well :) as a serial founder, I find AI exciting because after my 3rd company, thinking of my 4th was quite exhausting but AI has re-invigorated by ambition to start another company.

Exactly, keep adding to it - and they all add up.