Working as a feeelance consultant means you have to do marketing AND sales. (and backend paperwork as well). You need to be able to float through stretches of no work, and you need to be able to deal with clients who won't pay you.
Your product is yourself, so you start with brand building. What are your differentiators? (human) Networking is the most common way to market your services, but some write books, speak at conferences, have a substack, and blog too.
Setting rates and closing sales is another challenge. There are whole schools of materials to help with this.
Lastly remember you are trading your time for money. Your time includes the marketing, sales, and finance/taxes/billing. You may need liability insurance as well. With all that said your time is finite and not scalable - even if you charge top dollar there is a ceiling on how much you can make. Don't expect to get rich in this line of work by itself. (Side note: "ownership" - real estate, stocks, intellectual property, etc - are the scalable wealth builders)
I went down this route for a while, but ultimately decided I would rather just do the technical work and leave the rest to others.
The human brain does not have perfect memory. It is not always logical. And more often than not it is motivated and influenced by "external" forces - health, hunger, sex drive, environmental conditions, luck, spiritual inspiration, or whatever. The perfect worker is purely logical and has perfect memory and no external influences - never gets hungry or sick or wants to be the boss themselves. The AI race is funded by folks interested in creating the perfect worker, not a human. I have to agree with the conclusions of this paper that they won't be able to make humans. (But they don't really want to.) The Vatican has also published interesting works on this idea. The question is - if you take out everything that makes it human, can you call it intelligent?
Getting a sql query to optimal performance is still much more of an art than a specific science. Having the LLM generate a query that appears to work (correctness issues aside), is much more likely than the LLM generating an optimal performing query. While this may not matter for one-off queries common in analytics, when you start worrying about scalability, even the tiniest tweaks can make a huge difference.
Several hotels I've stayed in recently charged an "amenities" fee. When I asked what amenities one said it was for the free wifi and use of the lounge.
I'm another long time hackernews reader and banjo player. Back in the days when we used to work in data centers, I was pulling a long late night for some server maintenance with my banjo and I heard the night security guard come in on his routine patrol. I played a bit of dueling banjos (from behind a rack). He spun on his heels and got out of there. He didn't come back through again that night.
I worked with a devops person who moved everything we had set up in other regions _to_ US-East-1 because that is where you are supposed to run stuff. According to him, the other regions were just for DR stuff.
The Project, not Intel. It is Intel Project motivating the improvements which are making it possible for AWS, the box stores, the fast food joints, Google, and others to move in too. Intel the focal point for the money from the local governments, the state government, and the federal government to help make this happen.