fbrncci

10 ms·

91.
by fbrncci 3y ago
That's just the wrong tool for the job. I am quite heavy on the use of LLMs for my own day-to-day work (I max out 25 GPT-4 calls in 3 hours at least once a day). And I can't get anything decent out of Google Bard either.
92.
by fbrncci 3y ago
Well to each it’s own. I want to hack together stuff and make it work. The less time spend on code, the faster I can figure out what does and not.
93.
by fbrncci 3y ago
I can't wait for these concepts to work better, in a few years from now, I could imagine myself mostly prompting, making small code and integration changes, and writing 90% less code than I have been doing throughout most of my career.
94.
by fbrncci 3y ago
It depends! For example with Django, if I want something, which I already kind of know, but haven't done in a while, I will always go to 3.5 first, because its much faster, and I can get the answer within 5 seconds. But if there is any
95.
by fbrncci 3y ago
Life is good, but I need a sense of impending doom to keep constantly progressing, improving and growing. Not doing that would drive me crazy.
96.
by fbrncci 3y ago
I'm 5 years into my software development career, for now I am fine. But I am also making sure that my next possible job is already lined up. I'm not taking any chances with a single client, company or even a technology holding a d
97.
by fbrncci 3y ago
I'm in Asia, as a independent contractor and I work with and consult businesses in Europe. Currently that is pushing me towards €110k for the past 12 months. It's helping that I don't compete as outsourced "foreign talen
98.
by fbrncci 3y ago
Well surely I can't convince you, feel free to build the next AI startup on OpenAI then, and stop caring about any possible competition out scaling you once token limits on open source models become more in line with the walled garde
99.
by fbrncci 3y ago
Yes it’s against the ToS of most service platforms.
100.
by fbrncci 3y ago
Even if it was around 0.10$. This does not scale, it would need to be less than 0.01$ per generation to keep up with open source models where the cost effectively is 0$ (leaving our hardware). These open source models are still not replacin
101.
by fbrncci 3y ago
Those startups will move on to open source models because OpenAI api calls with 32k token contexts are way too expensive.
102.
by fbrncci 3y ago
When I started out as a freelancer / independent contractor, I made the conscious decision to not use any of these platforms, out of fear for my client base or finances being tied up and taken away from me due to a simple ToS violation
103.
by fbrncci 3y ago
But wait, there is more! In Langchain you can build constitutional chains ontop of your chains, to validate if the answer was really helpful, by doing just one more API call, with a new prompt, asking if the answer answered the question bas
104.
by fbrncci 3y ago
I think that is a phase, which will come to pass, eventually most of the plugins and apps which are going to be popular, will likely run their own models or use open source models. Because the API calls for complex applications to OpenAI ar
105.
by fbrncci 3y ago
It's more or less an advanced personal project to stay on top of the LLM learning curve, rather than just being exposed to news and press releases. I also have an appetite for further wrapping my mind around all of this. I already work
106.
by fbrncci 3y ago
I work with Langchain on a daily basis now, and so often I find myself asking; do I really need a whole LLM framework for this? At this point, the assistant I am writing, will likely be more stable rewritten in pure Python. The deeper and m
107.
by fbrncci 3y ago
No problem! You could probably improve it by fine-tuning GPT models on different categories of documents, prior to doing the vector retrieval from embedding. Fine tuning isn't available for GPT-3-turbo or GPT-4 yet, so I am waiting to
108.
by fbrncci 3y ago
Pretty easy, take a look at Langchain tutorials on YouTube. Basically you give it a set of documents, split these into smaller documents and then store these in a vector database and create embeddings (OpenAI, Jina, etc). Then when you inte
109.
by fbrncci 3y ago
Yes, the results are super customized if you sift them through a VectorDB first. I haven't tried fine-tuning but once I have everything down to a more stable system, I will try a more hybrid approach of both finetuning and storing hist
110.
by fbrncci 3y ago
I would, but I am afraid that everything in this sector is moving so fast. In 6 months from now there probably is a new approach or framework of doing everything how I am doing it right now.
111.
by fbrncci 3y ago
In an effort to keep myself on top of the latest development when it comes to AI and LLM's, I set out to work on a personal assistant, which serves as an replacement to journaling for myself. Throughout the day, I spend an hour or so t
112.
by fbrncci 3y ago
I got a contracting gig from one of those posts, pay is really great and I am up for my second year renewal. The interviews I had from those threads (two dozen), where also much better than any interview I got lined up from LinkedIn.
113.
by fbrncci 3y ago
LLM's are about as close to AGI as an apple is to a fully matured apple tree.
114.
by fbrncci 3y ago
This thread makes it sound like people are overestimating how good the content is supposed to be. Most clients want whats in their head, in writing, and they do not fully comprehend or care when its written badly (they aren't writers),
115.
by fbrncci 3y ago
I would say quite well! I'm currently spending 1-2 hours with Langchain and trying different approaches. I'm using a RecursiveCharacterTextSplitter with 1000 word chunks and 200 overlap, which may not be the best way of doing thin
116.
by fbrncci 3y ago
I'm currently creating a lot of private chatbots for people and organizations I follow online. I am scraping their content, from blog posts to articles and videos, and then storing these in vectors. Not for any commercial uses, but to
117.
by fbrncci 3y ago
I feel like I spend the entire pandemic, trying to understand my mind and my place in the world better. Than 2022 was the best year of my life, in all aspects except perhaps social one. Then towards the end of 2022 and now in 2023, I am tre
118.
by fbrncci 3y ago
I see your concerns, and I find them very valid, am I am happy that things are this way. I think you are right about some things, but probably very wrong about the outcome here. Software development is about to enter a philosophical stage,
119.
by fbrncci 4y ago
I would say that since LLMs made it on the scene my code output efficiency went 2-3x. As a backend dev I ended up implementing several frontend features at work, stuff I would have not touched previously. I did not blindly implement them. B
120.
by fbrncci 4y ago
I was earning similarly in Europe. The moment I realized how much tax I was paying (51% bracket), I moved to Asia to work remotely. I honestly don’t care what’s being paid from my taxes, but to me that amount felt like theft. I now pay ~20%

More ›