Digitizing and organizing old document scans for birth, marriage, and death records would be a huge win for genealogy research. The Mormon church would be a great customer for you.
Further, there’s guides and even coaching to help.
I have no affiliation with them other than as someone who has benefitted tremendously from everything on offer.
Definitely treat this like an addiction. If something, chemical, behavioral, or otherwise is causing your life to suffer, it’s an addiction.
I would guess your child gets more positive feeling from the game than life.
It can take a little time, but there is hope and a healthy way through.
On a related note, my college roommate was exactly the same way. He failed out of college. He worked for a very large game studio by starting at the absolute bottom and then eventually headed a new game category of the type of game he frequently played in college.
He now runs a game studio of his own making multiples of what I do.
As parents, it’s hard to watch a child take a different path than you would expect would be optimal for them, but sometimes they can surprise you by getting out of the local optima you find yourselves in and branch out to something new.
Good luck to you. From someone who previously suffered this same thing, I wish you all the best.
Whisper works ... kinda. I'm hoping there's another set of models released at some point, the error rate isn't appalling to me because i am transcribing TV shows and radio shows for personal use, so it's not mission critical.
There are a few whisper diarization "projects" but i've never been able to get it to work. Whisper does have word-level timestamps, so it should be simple to "plug in" diarization.
I don't need an LLM or whatever this project has, but i will see if it's runnable and if it's any better than what a couple podcasts i listen to use.
edit: see some people mentioning whisperx, which is one of those things that was cool until moving fast broke things:
>As of Oct 11, 2023, there is a known issue regarding slow performance with pyannote/Speaker-Diarization-3.0 in whisperX. It is due to dependency conflicts between faster-whisper and pyannote-audio 3.0.0. Please see this issue for more details and potential workarounds.
which means that what i gain is a ~3x increase in large-v2 speeds but i instantly lose those gains with diarization, unless i track down 8 month old bug workarounds.
I'll stick with the py venv whisper install i've been using for the last 16 months, tyvm
Amazing. I’m looking at building an app to look over the employment sections of the legal code and come back with results if things are allowed or not.
Answering questions like:
Can my employer do X?
As an employee in this country, what’s my minimum days off I can take?
And so on.
If your claims are true, then this will be exactly what I’m looking for.
I think spRAG should be pretty well suited for that use case. I think the biggest challenge will be generating specific search queries off of more general user inputs. You can look at the `auto_query.py` file for a basic implementation of that kind of system, but it'll likely require some experimentation and customization for your use case.
No. Changing your identity is much more powerful and sort of operates in a background process whereas willpower is more of a conscious foreground one and very difficult to sustain long term.
Read the book. It goes into some detail.
I’ve worked with thousands of people personally over the years to help them with supposedly difficult to treat addictions for instance. Willpower is fine to start, but completely fails in the long term. Identity change is the only thing I’ve found that works long term.
Run a few test theories and see what resonates. Are you the kind of person who goes to the gym every day? Are you the kind of person who works on their hobbies at night? Are you the kind of person who binges Netflix between dinner and bed? Are you the person who cooks dinner from scratch, or orders takeout at every chance? etc. I think of them a bit like stereotypes but applied to myself, and see if I think I fit.
It’s not supposed to be descriptive in this case, unless it’s a reflection or self-assessment. For goals, it’s supposed to be aspirational, like a vision you cultivate for yourself. You can meditate on that self-image, or identify other people with that trait and associate yourself with them, etc. The idea is that you will violate your sense of self by acting against how you expect yourself to be.
I started a company with a combination of no-code tools and LLMs to help me program some of the rest.
I’ve got a degree in CS, but haven’t programmed professionally in 15 years.
These new tools have allowed me to create something even larger teams would struggle with on my own in a matter of months.
I’m astonished at the progress I’m able to make without much assistance.
There is often something I’ll need to sit and debug or figure out the name for if I want to ask Google or the LLM helping me code, but that’s about it.
As far as I’m concerned, that means I haven’t needed to hire at least 2-3 full time engineers, be fully bootstrapped, and profitable from day 1 at launch.
Maybe I’ll hire people in the future, but so far the LLMs have replaced engineers, copywriters, designers, and customer support roles. I can handle it all on my own or have the LLM (in the case of customer support) do a decent enough job that I don’t even have to do it anymore.
But it didn’t really “replace” anyone in this case - you started a new company that might never have existed, and it might only be economically viable because you didn’t need to hire those people. Voila! We have a net increase in employment thanks to ChatGPT.
I think that's a great point. Though now we need to determine whether the result is a company that wouldn't otherwise exist, or a company that would exist but with lower profit margins. Also, would be good to know if this new company can undercut competition, which would likely mean that competitors had to find similar efficiencies.
It was certainly economically viable without LLM help, but the fact that I didn’t have to give up equity, get a loan, or use substantial amounts of my own funds to get it done at what would normally be a longer timeline to boot seems like a loss for some bankers and some freelancers or cofounders I would have needed otherwise.
I do agree that I’m creating something from whole cloth here so it’s not technically replacement as I haven’t let anyone go, but I’m going with the spirit of the question here.
I find it difficult to combine no-code tools with chatgpt. No-code tools are often no-code and point and click. Then chatgpt has to generate point and click instructions. What is your workflow?
I don’t use LLMs for this currently. Maybe one day. I’d certainly be happy if that was the case.
I’ve used tools like https://www.replo.app/ to replicate Ecommerce stores and use LLMs to rewrite copy though. That’s automating a huge amount of work.
I’m using LLMs to help me code backend coding tasks that I need done for backend services and API generation and interfacing.
So in this case, it's negation of 2-3 hires, and likely negation of using freelancers as well. I'm curious to know which LLMs you're relying on for which tasks.
Varies at the moment as I’m onboarding users slowly, testing, and iterating. About $15-30k MRR usually. Only been running a few months so it’s hard to extrapolate too far.
