Viewing profile — luckyt

I went for NextAuth - the use case was relatively simple, and I wanted maximum control.

Clerk has quite a few dark patterns in their free tier, eg: if your app is on Clerk free tier, all your users will be forced to log out and re-login every 7 days (and they try to o…

There are several issues that make the KV cache as-is unsuitable for caching across requests. First, it requires the cached tokens to be in the exact same position in the sentence,…

Yeah, I had a similar experience with Chroma DB. On paper, it checked all my boxes. But yea, it's alpha software with the first non-prerelease version only coming out in July 2023 …

It would be great to see more innovation like AI in DAW tools, but there are some challenges. The main constraint is it needs to process in real time, allowing just a few ms to pro…

I had the same thought, but overall, there is probably an order of magnitude more people using LLMs in applications or fine-tuning them compared to those trying to pretrain LLMs fr…

I guess this goes to show how challenging it can be to implement transformer neural networks correctly. There are so many ways in which you can make mistakes at various steps, and …

> I hope we find a path to at least fine-tuning medium sized models for prices that aren't outrageous It's not that bad; there are lots of things you can do with a hobbyist budget.…

Yea, ONNX runtime is mostly used for inference. The requirements for training and inference differ quite a lot: training requires a library that can calculate gradients for the bac…

I found it helpful to start with CUDA on numba since it lets you write GPU kernels in python. Assuming you're like most ML engineers and you're more familiar with python than C++, …

Yea, this was my experience too when I tried it out last week for my side project. It's easy to get started, but it's quite complex and disorganized and poorly documented. There ar…

Hmm, so is your conclusion that dog walking startups are not venture scale? But there are several dog walking startups that have received VC funding, and at least one that's IPO'ed…

Yeah, it's not clear how to estimate the market size or TAM of an idea even if there already is a market, like take random examples of a dog walking or podcasting app. You can easi…

I think the author's point broadly still holds -- you can get further with more engineering resources and data, whether you're using 2015 era models or 2023 retrieval-augmented LLM…

There's definitely a lot of opportunities for technical debt in machine learning projects that don't exist in usual software development, which makes careful design decisions even …

It's really interesting that XGBoost failed when ran on a dataset that had no noise. I've also seen a similar thing occur with the Adam optimizer when training neural networks on p…

Ah, I failed to consider that. The original post is correct.

This is incorrect: there's no reason to expect that X and Y will each appear 3 times in 6 trials if their probabilities are equal. If all 3 measurements of X are smaller than all 3…

Yes, ideally you would carefully read the paper and evaluate its methodology, evaluations, and results, but this requires you to be knowledgeable in the field. That's why peer revi…

You're absolutely right. A good way to tell if a paper is legit is search it on Google Scholar and see how many citations it has: this paper has none. If it's an arXiv preprint (no…