(Which to be fair does fit some definitions of “few people”)
1,361 karma · joined June 21, 2014
https://delphihq.com
(Which to be fair does fit some definitions of “few people”)
They use it to transform raw data in their data warehouses into views/tables that people can use for analysis.
Long term they’re really worried about AI alignment and are probably using this to understand how AI can be “tricked” into doing things it shouldn’t.
(Full disclosure: I have a side project exploring something similar. Demo: https://hermes-odmf.onrender.com. See the Metrics Catalog link for available metrics; all data is from 2018)
(This is a genuine guess, not a knock on YC)
It was a little surprising since I’ve only ever seen the Spanish kings referred to as Juan Carlos and Felipe in English-language media, never by translations.
(My email is on the website in my bio, or ping me on the dbt slack if you’re there).
hfsp
https://en.wikipedia.org/wiki/List_of_countries_by_life_expe...
I never used Second Life, but from that video the metaverse does feel a lot like ActiveWorlds circa 2007.
She’s being replaced by Javier Olivan, who has been at FB forever and is the head of their Growth org.
Step 2: paywall customers’ access to their own data so you can brand it as a premium feature/product
Step 3: ???
Step 4: profit
I consider myself an EA and donate to the normal GiveWell causes (mosquito nets right now, I think), as I’d suggest other small donors do. Your marginal dollar won’t make a difference to extinction risk research, and certainly not if you give it to AI research.
But I wonder if you are trying to do too much. For example, why would I use Bach to transform my data instead of just doing it in my dbt project or wherever else I handle warehouse transforms?
I would amend #1 to “the adherents of the conventional view are convinced its truth is beyond question.”
Truth is actually really important, maybe the most important thing for heresy. A defining feature of heresies is that those enforcing them believe strongly that the heresy is untrue.¹
Take PG’s example of doubting the divinity of Christ. Truth isn’t secondary in that case - just the opposite. Believing it, literally, is the core feature of Christianity.
> The clearest evidence of this is that whether a statement is considered x-ist often depends on who said it. Truth doesn't work that way. The same statement can't be true when one person says it, but x-ist, and therefore false, when another person does.
This is a fallacy. Insofar as this phenomenon exists (less than people think), it’s because the belief content of the statement is different depending on who says it. There’s a common view on the identitarian left that “minorities can’t be racist because racism is about power and they don’t have any.” But a Latino professor who spends his time arguing that Black people are genetically inferior to other groups will get cancelled just as quickly as a white one.
¹Or need to be seen as believing it 100%
What’s the quick pitch for why I should use this instead of Cube or dbt’s metrics layer?
I was working on a little API project this weekend[0] and struggling with how I was going to do authentication. Then I saw this post, and it took me about 10 minutes to add Kable and have it 100% working in my code.
Nice work!
I'm curious how you differentiate yourselves from Airbyte, which isn't really designed for reverse ETL but can be used for it. And do you ever see Castled supporting regular ETL?
Right now there is a lot of separation in the market between ETL and reverse ETL, but it seems like a pain to maintain separate tools when you could just do both in one.
Would be happy to chat and compare notes if you'd like!
We're still early enough in the data world that it's easy to find that sense of ownership over the whole "data product." You probably won't get paid as well, although it's possible, but the pay won't be bad either.
Maybe I am missing something obvious, but who is he talking about here?
If so, it would make sense that for the average franchise, doubling salary wouldn't be viable. But surely there's a distribution where some stores are much more profitable than the average, and for those stores it would. Presumably that's part of the reason why Denmark only has 80 McDonalds (about 1/3 per capita of what the US does[0]).
[0]https://en.wikipedia.org/wiki/List_of_countries_with_McDonal...
In general I suspect AI tools like Codex will have a large impact on other technical roles like SWEs and DBAs, but I won’t try to predict exactly what that will look like.