larve

14 ms·

121.
Review of Lakoff's “Metaphors we live by”(norvig.com)
2 points by larve 4y ago|0 comments
122.
Linguistics and Programming Languages (2006)(250bpm.com)
41 points by larve 4y ago|9 comments
123.
The GPT Quine(the.scapegoat.dev)
3 points by larve 4y ago|0 comments
124.
LLMs will fundamentally change software engineering(the.scapegoat.dev)
7 points by larve 4y ago|0 comments
125.
Claims That AI Productivity Will Save Us Are Neither New, nor True(cigionline.org)
2 points by larve 4y ago|0 comments
126.
Decision Transformer Interpretability(lesswrong.com)
2 points by larve 4y ago|0 comments
127.
Delivery Drivers Are Using Grey Market Apps to Make Their Jobs Suck Less (2021)(vice.com)
1 points by larve 4y ago|0 comments
128.
Just Ask for Generalization (2021)(evjang.com)
46 points by larve 4y ago|1 comments
129.
A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT(arxiv.org)
6 points by larve 4y ago|0 comments
130.
We can talk to AIs – an overview of the research of the last 10 years(medium.com)
2 points by larve 4y ago|0 comments
131.
LLMs – a paradigm shift for the pragmatic programmer(the.scapegoat.dev)
7 points by larve 4y ago|0 comments
132.
Thoughts about the ethics of Large Language Models(the.scapegoat.dev)
1 points by larve 4y ago|0 comments
133.
Prompt engineers: tech’s next big job?(washingtonpost.com)
3 points by larve 4y ago|2 comments
134.
by larve 4y ago
I wouldn't care at all if IBM's CEO was actually an AI. But I sure as hell care that the person I have on the phone when inquiring about something is a human.
135.
Zygomys – a Lisp implementation in pure go(github.com)
1 points by larve 4y ago|0 comments
136.
by larve 4y ago
A lot of what this takes are missing is that the unreliability of those models is not a hurdle for a tremendous amount of applications. If there is a human in the loop that can now write a couple of words in a fuzzy human language, check or
137.
by larve 4y ago
Having lived in some of those scenes, the environment or the identity or political affinities that lead to the creation of those groups are only indirectly linked to the cultural aspects that emerge from it. There very much is specific musi
138.
by larve 4y ago
Cigarettes, 100%. gums are not fun and really just help avoid the withdrawal.
139.
by larve 4y ago
I was replying at that level to echo the other comments. It's reacting to the "they merely chose to do so" part. It might not be the best parent to reply to, but it's what spawned a whole thread I wanted to sibling.
140.
by larve 4y ago
I think most issues brought up here would be addressed by writing rfcs and have those go through a review cycle, potentially with attached code that is clearly understood to be a proof of concept. It leverages async by giving someone the au
141.
by larve 4y ago
This narrow view of addiction is also missing the fact that smoking has many positive effects. I smoked from age 13 to age 38, and I’m pretty sure smoking made my life better overall. Besides allowing me to socialize more easily, it’s an in
142.
by larve 4y ago
Maybe BertVIZ? https://github.com/jessevig/bertviz
143.
by larve 4y ago
While I’m not in the consulting business per se, that book is a goldmine just about the topic of giving advice as well. Highly recommended (as is anything by Weinberg)
144.
by larve 4y ago
Mathematics only clicked and became fun for me when I started using Wolfram Mathematica, because I could fairly easily mess around with the formulas I saw in books until I understood the types and arguments and what is an index vs a referen
145.
by larve 4y ago
Yeah, I think my point is more that this makes it possible to iterate quicker while maintaining quality, which might be beneficial. Of course if you are working in a team that cares about the result (and the business does too), then iterati
146.
by larve 4y ago
It depends on what you mean by beginner, but I would say for someone who has been futzing with a couple of js boot camps or building some php websites, it’s what takes you from hacking to software development. Completely new to programming,
147.
by larve 4y ago
Peter norvig’s paradigms of artificial intelligence programming, despite its age, is a delight. https://norvig.github.io/paip-lisp/#/
148.
by larve 4y ago
I wouldn’t underestimate the ability to iterate on fully finished features quickly, if you have to operate with end users. Closing a feedback loop can lead to threshold effects.
149.
by larve 4y ago
Seeding gpt (or using copilot) on snippets of your code is extremely effective. I often paste a function of my codebase before asking code questions, as in: “here is an example of a product class, generate the swagger spec for a user crud a
150.
by larve 4y ago
I haven’t had too many issues. I must admit I use the api and copilot a lot.

More ›