Yes, but then a PhD is a chicken and egg problem. You can't get into the top PhD programs under a respectable professor - unless you have some publications under your name.
So how does one get published, enough to get into a good PhD program?
Right. I think the ability to run tests/compilation as a feedback mechanism is definitely interesting .
I wonder whether the problem
is that its landed on an incorrect output sample (i.e re-sampling another code output would fix it ) .. vs … a feedback loop on the compiler output is what it really needs ..
Precisely. I think it helps with smaller/mundane tasks (that it has seen in its training), but the tasks that actually require a higher level reasoning and understanding of the bigger picture - are not something we can expect the current LLM's to do.
After using code LLMs' for a bit, I wonder whether there is significant productivity gain in using them. I found that its limited for me.
Main reason:
a) to find the right code, I have to keep sampling it, and b) it doesn't seem to be able to solve larger / more complex problems that I actually find more need for.
I found some interesting research on combining planning-algorithms for complex problems, and some ideas on guiding the LLM's decoding process towards correctness by optimizing it via reward functions and reducing the search space.
I've detailed and summarised the main points in the post above.
Questions:
1) Do you find code LLM's really useful? Please share some stories / examples where they help vs they didn't. I'm trying to form a better understanding they are just fancy, vs actually productive and useful to most
2) Any other research ideas being pursued in this field ? / what are you trying ?
yeah, I think I like your approach, but the multiplayer aspect seems to be a great add-on, as opposed to a must-have feature to get the product started.
Both Go and Rust, and wasm are new to me - so i'm wondering about whether one is superior to the other.
I think figma has a lot of feature, and penpot seems a great replica of it, but I'm not too sure if the SVG + Clojurescript is a great way to build it..
I'm really looking for the core figma feature set to build.
1. Ability to have a canvas with zooms / freedom to create elemnets
2. Components and instances, and tree structured layers
3. Autolayout + CSS options. Frames and Pages
I think with the above 3, there's a viable basic feature set that would get one off the ground.
I did try the figma plugin apis, to extend native figma capabilities - but there's a limit to things you can do with it and it's getting to a point where more investment in the figma plugin seems a bit wasteful
Yeah, PenPot (mentioned in the comments below), seems to have chosen SVG as their choice of rendering.
But, looking at a few customer reviews - the performance issue seems to be noticable with the SVG approach (zooms are not clean, and more than 20 screens it starts feeling slow).
What would your customers need to make them want to pay for it ?