cjonas

15 ms·

181.
by cjonas 2y ago
Anyone got the insider scoop on how the "easy" mode of the delta chess app ended up being a savage?
182.
by cjonas 2y ago
Favorite: earthworms Least favorite: spotted knapweed
183.
by cjonas 2y ago
By the time you've figured out how to install and run this, you've probably learned more than the class your cheating in had to offer.
184.
by cjonas 2y ago
Ah yes, we apparently would prefer an oligarchic theocracy.
185.
by cjonas 2y ago
If you're not an expert in something, you should probably listen to those who are... The general public does not have the time to study climate models and review raw data on vaccine efficacy. Without overall "trust in science&quot
186.
by cjonas 2y ago
It's not an advanced algorithmic feed designed with the sole intent to consume as much of you attention as possible...
187.
by cjonas 2y ago
Vet cool! I've always wanted something like this in my note taking (currently use obsidian), but it being a "webapp" is kinda a deal breaker. I want something that runs and stores my data locally.
188.
by cjonas 2y ago
Did you mean "retrieval is comparatively inexpensive"? I think I'm on the same page but this threw me off.
189.
by cjonas 2y ago
I never said any of this will be good for society... In fact, I'm confident the current trajectory is going to cause wealth inequality at an entirely new level. Underestimating the impact these models can have is a risk I'm trying
190.
by cjonas 2y ago
The difference is a junior cost 30-100$/hr and will take 2 days to complete the task. The LLM will do it in 20 seconds and cost 3c
191.
by cjonas 2y ago
Closer to "bingo" than a proper skill tree. I was expecting something more like this: https://roadmap.sh/
192.
by cjonas 2y ago
I think typescript benefits even more than JavaScript because the type signatures add so much context. After defining the types and signatures copilot will often complete the body within a few changes of perfect
193.
by cjonas 2y ago
I just ported 10k loc of react classes to function components using gpt-4o. The changes are mostly trivial, but would be fairly time consuming and tedious to make. It took me a few hours instead of a few days.
194.
by cjonas 2y ago
Could solve this with a migration framework (I'm sure there is something for sqlLite). I've also done something similar with object/document storage. Store the version of the schema in the record and write a map function for
195.
by cjonas 2y ago
I don't disagree on a conceptual level, but on a regular basis I deal with companies completely lacking any real access model, users without MFA, blanket admin level access, etc... getting sold on this particular product and something
196.
by cjonas 2y ago
Gotcha... so basically encryption of disk at rest prevents someone from walking out with a drive... Encryption "at rest" in the database prevents someone with server or direct db connection from pulling the data. I had never reall
197.
by cjonas 2y ago
Sorry, still reading so having gotten there yet. How does Salesforce offering a "light" version of encryption at rest improve security? Or are you saying it's a better balance of performance / security by only selective
198.
by cjonas 2y ago
Salesforce has an addon product called "Shield" that cost 30-50% of your overall license cost. It allows for encryption at rest at the field level (not all fields are supported and introduces query limitations). Companies purchase
199.
by cjonas 2y ago
I agree, but would say it different: > AI will be a perfect programmer in the future for all NON-niche areas There's going to be a positive/negative feedback loop that makes it hard for new languages and frameworks to gain popu
200.
by cjonas 2y ago
Ya if it's GPT-3.5, I'm actually surprised the accuracies were so high! I've been pairing with GPT since 3.5-turbo. I run 20-100 queries a day (have an IDE integration). The improvements for GPT-4 over 3.5 are significant. S
201.
by cjonas 2y ago
I scanned the paper and it doesn't mention what model they were using within chatgpt. If it was 3.5 turbo, then these results are already meaningless. GPT-4 and 4o are much more accurate. I just used GPT-4o to refactor 50 files from r
202.
by cjonas 2y ago
I was able to fly the DJI avata in full manual mode with maybe only 4 hours in a sim. I'm not sure if that's because it just "clicked" or if the avata is just way easier than other FPV drones. I definitely think the abil
203.
by cjonas 2y ago
All the game changing AI programming I've seen is coming from LLM... Regardless, there will always be a lack of training context for newer languages or frameworks.
204.
by cjonas 2y ago
Does your company not just use an automatic formatter? Set a prettier config, format the entire codebase and never have to deal with another formatting change in a PR ever again.
205.
by cjonas 2y ago
New languages are at an even bigger disadvantage now with the rise of generative AI programming. A fancy new framework that is objectively better may actually be less productive because AI haven't been trained on it
206.
by cjonas 3y ago
Am I missing something is or Salesforce (the og SaaS) missing from your list?
207.
by cjonas 3y ago
Imo it's unfortunate that python is the dominant tech of this domain. Typescript is better suited for the inference side of things (I know there a ts version of most things, most most companies are looking for python dev)
208.
by cjonas 3y ago
I think the disconnect is that he's talking about building plugins/"gpts" inside of chat GPT while others are thinking about using the API to build something from scratch?
209.
by cjonas 3y ago
The "as code" means the quality and stability of the implementation varies wildly. If you use a declarative style with proper SDLC in place, then it's fantastic. On the other hand, I'm currently reviewing a project usin
210.
by cjonas 3y ago
Would love to see it. How do I get in touch (preferably without ending up as a marketing lead in your system)?

More ›