getmeinrn

The OP said "This is a pretty low bar request IMO" suggesting that the problem they expect an LLM to be able to do is not the hard problem you're saying Google and Yelp has not solved. It's a different problem.

That's mostly safe, but even then, a user could execute "SELECT SLEEP(100000000)" thousands of times and DoS your database. There are other unsafe functions that a readonly user can execute as well. I've written extensively on some of the attack surface here https://docs.heimdallm.ai/en/latest/attack_surface/sql.html

HeimdaLLM can allowlist functions and constrain queries to ensure that required conditions exist. This makes LLM + database usage have far more utility, for example, a user can be restricted to only data in their account. Support for INSERT and UPDATE is coming very soon.

The hard part is the SQL query, because you need to make sure the SQL query is safe to execute. Collecting data is far easier by comparison, but you absolutely could use an LLM for that too.

>I can't even get ChatGPT (+ web plugin), when given a list of restaurants in NYC, to tell me which ones are still open vs have closed, and what their hours of operation / locations are.

This is doable using a tool I've built. The key is to have that data in a RDBMS and to use an LLM to generate the SQL query that answers your question. Companies haven't offered this yet because there's no safe way to execute these queries on your behalf. Which is where my library comes in[1].

1. https://github.com/amoffat/HeimdaLLM

The entitlement and audacity of people who consume open source blows me away. I've maintained a project for 12 years and recently someone wanted me to help them implement the software in their system. I politely told them that since this wasn't a bug, they would need to purchase a support package. They then accused me of trying to "sell open source software" and closed the issue. People are unbelievable. Fuck me for trying to make a living providing you personal development time, using my software that I've supported for free for over a decade.

I think we're just scratching the surface of apps. As we figure out how to integrate this technology in novel ways (not just "here's my app + AI!!!!"), it will open new doors.

Shameless self-promotion, I'm trying to build some of those intermediary pieces. I have authored an open source library[1] that lets businesses externalize LLMs to their users, so that users can use natural language to query their data in your database. The goal is to try to simplify UIs to have more natural language components, without needing to send your data to an LLM.

1. https://github.com/amoffat/HeimdaLLM

I'm surprised that a search engine that isn't powered by AI is able to raise money. From the direction things seem to be going, LLM tech may very well level that entire industry.

Ah, they edited it. Must be why I'm eating downvotes now :)

Might want to rethink posting about committing insurance fraud.

>Employees, having tasted the fruits of flexible work, felt averse to relinquishing these newfound freedoms.

Translation for sociopathic leaders: never let your workers taste something good. It only took a global pandemic that forced people home in order for us to experiment with something that a huge portion of the population sees as beneficial to their well-being. Where were the small scale experiments before the pandemic, and why did they never catch on?

Makes sense to me, and if SBF was as smart as he thinks he is, he should have thought of this possibility. Why would an individual think they have any authority to enforce an extradition agreement made between two countries? Completely nuts.

I assume they said "government assistance" because they are limiting the idea of government spending to just giving money to help poor people. If they had said "government spending" in general, their point would be indefensible.

"Our team's lies are good, the other team's lies are bad."

Something I've learned over my life is not to automatically trust that someone has done or will do the right thing just because they stick their neck out. Sometimes they stick their neck out because they don't actually understand the risks and consequences, but they are pretending that they do. They're artificially signaling confidence.

In this case, the CEO risking his life on this third(!) voyage means 1 of 2 possibilities: 1) The CEO has done the due diligence and has a thorough understanding of the risks and the safety protocols involved to minimize those risks. 2) The CEO is naive.

"A programmer's perspective" but nothing unique to a programmer parent. I assumed it was going to be about all of the bad habits that go with being a dedicated tech worker, because there are lots of them.

I'm not taking a side here, but not every law has to extend to the general case for it to fix a problem. If we required that, there would be almost no laws.

>In reality, a smart business should be grateful that there’s someone that can get shit done in less than 8 hours, it means that they can do more for you in less time, and if they’re doing non-competitive side work, maybe you should pay them more so they don’t go and do that full-time.

In my view, the biggest crime of a manager is disconnecting my incentives from my work as an IC. If I create more value, it should be a manager's top priority to create the positive feedback loop of incentives. Because why else am I doing more, then? It should seem like common sense, but every manager I've had has not been remotely interested in creating that positive feedback loop. Instead "that salary is too far out of the band for this role and wouldn't be fair to everyone else, and you have to do X,Y,and Z to qualify for this other role, and your last self-review was too honest so I'm going to use it to negotiate against you..." I've even had a manager complain about my salary being too close to his salary, as if there was some requirement for his ego that I make less than him!

Every raise or promotion I've ever had has devolved into some form of "you're giving me this or I'm going to leave", and then 9/10 times they reluctantly fork it over. There is no good faith, only power. And power is getting what you want after calling all the bluffs. Managers are adversarial, and their role is to get you to do as much as possible for as little as possible, while at the same time swallowing their pride when their bluffs are called. I hate that it is like this, because it could be a lot better. But nobody seems to know how to manage.

Don't get me wrong, I get the position that managers are in. The upper management demands they get certain results, and if they don't do it, they get replaced. But it's still a failure of the manager for not communicating effectively how creating positive feedback loops with your top performers will make everybody money. Of course this depends on the upper management to actually pick valuable things to work on...

I agree the common sense around this issue seems to be lacking. I think this issue is more of a political lightning rod for those who "hate workers" vs those who "want government to control everything."

It's not an insane conspiracy when you acknowledge that they bootstrapped their community through fake accounts[1]. This shows they have no reservations about using deception to increase their brand's reputation. Fake app reviews falls into that category.

1. https://www.themarysue.com/reddit-fake-account-origins/

>for some reason

AGI

A dead golden goose containing one last giant golden egg. They don't care if it's dead, so long as they can cash out on the archived data that can be sold for training.

>I guess a caveat to that is that some people have to- and should- work to be generally likable. I have known some very smart people I would really prefer to never have to work with again.

I think you hit on something here. When someone is viewed as generally unlikable, excluding them is contagious. No matter how hard they work at it, it will always be socially easy and acceptable by the in-group to exclude them. There is relatively no pressure on the in-group to work at being more accepting towards "unlikable" people, like there is for those people to "work to be likable." People are extremely unforgiving in social interactions.

Yep, and if you've ever worked with someone who is a "dud" but leveraged their connections to get the position, you'll know how frustrating it is to realize you're competing with this outsized advantage.

I've been part of different hiring processes long enough to see just how many people get hired because they know somebody on the team, when there were other candidates who were as qualified or more qualified. When we can hire 1 person, my experience is that it is 1000% biased towards knowing somebody. Sure, if there is an undefined number of open reqs, you can argue that there is still space for the more qualified, unconnected applicants. But when you're dealing with a finite number of open reqs, there's actually not, and it is a zero sum game.

As a counter point, only 1 of the 5 tech jobs in my career have come through a connection. It can be done, and in the long run I think it's better for personal and professional growth to get opportunities that way. Not to mention that every time you get an opportunity through a connection, you're potentially screwing over someone else more qualified, without a connection.

Networking is an enormous time sink, where 95% of your interactions are unfruitful, but you have to pretend to be enjoying them anyways. Maybe you find one connection that is worthwhile, and of that one connection, you have to carefully nurture it, without giving off the "this is the one important connection" vibes, because people can smell that a mile away. And sometimes that connection evaporates under mysterious causes. It's pointless. I truly can't justify wasting my time doing it when I could be producing things that anonymous people appreciate and rely on. I'd rather serve them than waste my time playing delicate social/power games with people in my proximity that I don't even relate to. At least when you serve anonymous people, they're generally honest about the quality of your work. You can't rely on anyone in these networks to be honest with you like an anonymous person can.

Companies have the good faith belief that employees would be so outgunned in a legal fight that most of the time they won't even challenge invalid clauses.

I don't know much about the Unabomber's victims. Where can I read about your friend's dad?

You mean to pull a trigger, correct? There are plenty of other areas where AI can shine like (excuse the layperson terms, I don't know what these are officially called) threat identification, intelligence gathering, logistics, population control.

If you've gone without Reddit for this long, you're fine. The odds of Reddit having a net positive impact on your life are pretty low.