Magmalgebra

6 ms·

31.
by Magmalgebra 1y ago
As someone who used to teach, this does not worry me (also, they mention skill atrophy - inherently less concerning). Putting ChatGPT in front of a child and asking them to do existing tasks is an obviously disasterous pedagogical choice
32.
by Magmalgebra 1y ago
Well... yes? Essays are tools to force students to structure and communicate thinking - production of the essay forces the thinking. If you want an equivalent result from LLMs you're going to need a much more iterative process of criti
33.
by Magmalgebra 1y ago
Anyone concerned about cost should remember that those costs are dropping exponenentially. Similarly, nearly all AI products but especially OpenAI are heavily _under_ monetized. OpenAI is an excellent personal shopper - the ad revenue that
34.
by Magmalgebra 1y ago
Something that can be hard to appreciate if you haven't managed this sort of project is that it can be surprisingly hard to throw money at the problem. If you try to hire at your regular "bar" for skill for boring work like t
35.
by Magmalgebra 1y ago
> or are there still good reasons for using it in a new project in 2025 it's not clear there ever was. Most of the big users I'm aware of, like Stripe, don't seem to have needed it and regretted the decision. Big data did
36.
by Magmalgebra 1y ago
This isn't a "tech" thing, it's a "professional managerial class" thing. This same tests is applied in banking, finance, consulting, sales, or any number of other highly renumerative white collar professions. T
37.
by Magmalgebra 1y ago
> So you are saying it is better if the candidate lies. This is a toxic framing of an essential test. Constructing polite fictions is an essential skill for collaboration - no less essential than coding. Saying you're leaving in par
38.
by Magmalgebra 2y ago
I think experiences are highly variable based on the person and stack. I'd say I've gotten a ~50% productivity boost. I'm quite senior, make FAANG money. I work in typescript, rust, and go - mostly typescript. LLM coding is a
39.
by Magmalgebra 2y ago
Descriptions of tables is insufficient (we had it) - you also need descriptions of the systems writing to the tables. My favorite example was a report that was only accurate if generated on a Tuesday or Thursday due to when the ETL pipeline
40.
by Magmalgebra 2y ago
Unfortunately this doesn't address the problem I'm describing. My team had these ontologies available to the LLM and provided it in the context window. The queries were ontologically sensible at a surface level, but still wrong. T
41.
by Magmalgebra 2y ago
Ususally the analyst, but sometimes the exec - hard to miss when a report implies your revenue has shifted 90%+ in either direction since the last time you read a report :)
42.
by Magmalgebra 2y ago
I think this post underestimates how the degree to which “what data is correct” is deeply contextual. My team created an identical hypothesis to this doc ~2 years ago and generated a proof of concept. It was pretty magic, we had fortune 500