captnObvious

Ive mounted monitor arms to the front of the desk, rather than the back, and extended them out toward me for a somewhat similar situation. Bluetooth keyboard goes in my lap, thumb ball mouse goes on my arm rest. I can extend the monitor about 2.5 feet toward my face in this way. Hope it helps

For law firms creating their own repositories of case law, having LLMs search via summaries, and then dive into the selected cases to extract pertinent information seems like an obvious great use case to build a solution using LLMs.

The orchestration of LLms that will be reading transcripts, reading emails, reading case law, and preparing briefs with sources is unavoidable in the next 3 years. I don’t doubt multiple industry specialized solutions are already under development.

Just asking chatGPT to make your case for you is missing the opportunity.

If anyone is unable to get Claud 3.7 or Gemini 2.5 to accelerate their development work I have to doubt their sentience at this point. (Or more likely doubt that they’re actively testing these things regularly)

I’m surprised this thread isn’t seeing more comments, it’s got a little bit of everything.

You’ve legitimately never tested something you built in production, after having already tested it in staging and local? You’ve just, had complete faith in staging production environment parity your entire career on every project and it has never failed you?

I’m sorry man but I don’t believe you at all.

This is what I’m thinking. All of the kids I know from 16-22 are the most level headed group of young adults I’ve known. It is hilarious to me that this group of brilliant technologists leans so heavily towards seeing the absolute worst in every data point.

Could it be that, kids are doing less drugs because they’re more informed, less bored, and less reckless than previous generations?

We all aspire that our kids will do better than we have. We did our best to instill a sense of confidence and worth in them.

What if it is finally starting to just, f’ing work?

It sounded like an optional request not a command. I’ve managed and I’ve been managed and the relationship ships I’ve built along the way are deep enough to have conversations like these. We’re doing life with this people it’s not a dystopian novel about capitalism. If you have a genuine person as your boss moments like these help form friendships, and don’t feel like a dark Dilbert comic.

Friendships increase team communicability which bolster project success.

I agree with you entirely except for if we skip the year over year part.5-

Time of purchase upgrade ability, if we’re talking about getting to 128 or 256 GB of RAM. Time of purchase to upgrade to multiple high res screens that match. Dedicated GPUs… I bet there is a top of the line home hobbiest LLM oriented GPU from Nvidia or AMD in the next 3 years that will cleanly connect to recent chip architectures. I doubt it will run optimally tied to a Mac. It’ll be something that you could also rack in a server.

I think we’re reading too far into the authors impostor syndrome.

He’s making contributions in Rust already. His opinion isn’t invalid just because he has a bias, he opens by acknowledging his bias.

Why isn’t AGI ever going to happen? Ever?

Check out chatgpt code interpreter. One step down…

I’ve used GPT-4 as a pair programmer actively for over 40 hours now. Using it to do things you have 90% of the solution for is a bit of a waste of time.

Using it to help trouble shoot errors in logs when you’re feeling frustrated, and get a basic framework up from scratch to start modifying toward the your goal are the best use cases I’ve seen. The ability to get yourself unstuck in minutes is life changing. The biggest constraint right now is short term memory loss. It can nearly always recognize on a second pass through what mistake it made.

Keep in mind 32kb token requests are already in beta… and 4kb requests are what you have access to now. So… the memory thing is potentially a dead point soon. Though at 6 cents per 1kb if it actually has to volley 32k over to you it’ll almost cost $2.00 based on current API pricing.

That 32k option would allow GPT-4 to hold every non-monolithic project I’ve ever worked on in working memory. The pertinent parts anyway, it wouldn’t need to read CSS files or imported modules unless it really gets forced down a specific rabbit hole.

This will be one more reason new projects won’t be monolithic architectures. Easier for the AI to help out.

I hope y’all have a plan for when AWS decides to pick up your open source project and turn it into a managed cloud solution. It’s a pattern of theirs. And with the way egress charges are structured they’re likely to snap up any clients straddling their cloud and yours.

Holding onto inventory is death. In auto and housing sales if you’re a dealership you’re leveraged in order to hold that inventory (paying interest on it, less friendly interest than you get as a government backed mortgage borrower) and you’re paying insurance and maintenance on that inventory. Vacant houses depreciate very quickly.