I'm keeping an eye on Tenstorrent for this. Pricing seems like its going to end up being in between a super memory dense unified memory platform, and a purpose built GPU.
Definitely on the edge of what would make sense at home, but its interesting.
how exactly did you jump to the conclusion that American ?citizen? taxpayes are propping these two people up financially? They are presumably both paying income tax in the United States as well, probably a lot of it based on his wife's profession.
some models do ok with instruction, some don't. You still have to review pretty thoroughly and make corrections or re-prompt.
For the most part I use claude code w/ opus plan mode + sonnet implementation mode, and it does pretty well. I go through the planning process for anything other than the most basic changes.
> You go from formulating unique solutions to flagging things that look wrong and then just picking an alternative LLM generated one.
totally depends on how you use the tools. Plenty of people fall in to that scenario, but plenty of others explicitly tell the LLM tools what and how to build something.
A good generic terminal UI app for reviewing git diffs, making comments, etc...
Lots of these have started popping up, but almost every single one of them is a TUI interface for github, or gitlab.
What I'm building is for local git by default, but has an extensible plugin system to support integrating with github/gitlab/azure ado/etc... for their PRs and approvals, but forced in to a single consistent UX and workflow.
Its good enough for basic diff reviews that it has become my daily driver for about a month for reviewing my own stuff before I push changes remotely.
I kind of disagree. You are describing a kind person who is extremely valuable, a person who is proficient in SWE but also has domain specific skills in some niche.
That's great, but its nowhere near the norm, and people have been doing generalist software engineering for decades. There has been a sufficient amount of work for a long time to be performed by generalists that it has been a very reasonable career.
IMO AI is the first thing that has ever actually challenged that.
I wish it made sense to do residential solar where I am. It probably does technically, but i hate the idea of spending a ton on a system and then STILL have to pay my power company; if you are connected to the grid at all where I am, you pay the power company $5/kw/month of solar capacity and your excess sell-back rates are insanely bad (0.03/kwh, vs billed usage rate at $0.17/kwh)
I use it as my main platform right now both for work/swe stuff, and person stuff. It works pretty well, they have the full suite of tools I want from general LLM chat, to notebookLM, to antigravity.
My main use-cases outside of SWE generally involve the ability to compare detailed product specs and come up with answers/comparisons/etc... Gemini does really well for that, probably because of the deeper google search index integration.
Also I got a year of pro for free with my phone....so thats a big part.
the scaling and UI framework issues are by far my biggest pain point. I will inevitably end up with an app with tiny and/or blurry UI elements every few weeks and have to spend a ton of time figuring out the correct incantation to make it better.
This is on a pretty clean/fresh install of current ubuntu desktop
agreed. I haven't done LFS, but ive done arch and plently of other distros for a good while and I definitely wouldn't say I have a rock solid understanding of the fundamentals.