edublancas

5 ms·

1.
SanitAI: A reverse proxy to remove PII data from OpenAI API calls(github.com)
15 points by edublancas 2y ago|0 comments
2.
How do AI agents work, anyway?(blancas.io)
1 points by edublancas 2y ago|0 comments
3.
Smolagents, a simple library to build agents(huggingface.co)
2 points by edublancas 2y ago|0 comments
4.
Ploomber (YC W22) Is Hiring Engineers (Infra, Backend, Growth) and Ex-Founders(ycombinator.com)
1 points by edublancas 2y ago
5.
by edublancas 2y ago
TIL there is pgvector and pgvecto.rs
6.
by edublancas 2y ago
Cool work! I've been working on a similar product. Users can select between Streamlit/Shiny: https://editor.ploomber.io/ - so not necessarily for BI (although you can use it for that), but more broadly focused on
7.
Mosaic: An extensible framework for linking databases and interactive views(idl.uw.edu)
3 points by edublancas 2y ago|0 comments
8.
by edublancas 2y ago
has papermill deprecated the ipython runtime? I used papermill extensively in the past and I never saw that in their docs.
9.
by edublancas 2y ago
Papermill is great but has quite some limitations because it spins up a new process to run the notebook: - You cannot extract live variables (needed for testing) - Cannot use pdb for debugging - Cannot profile memory usage You can do all of
10.
by edublancas 2y ago
I'd say how much is good enough highly depends on your use case. For something that still has to be reviewed by a human, I think even .7 is great; if you're planning to automate processes end-to-end, I'd aim for higher than .
11.
by edublancas 2y ago
thanks a lot for the feedback! you're right, this is much better input data. I'll re-run the code with these tables!
12.
GPT-4o doesn't need HTML tags to parse content accurately(blancas.io)
2 points by edublancas 2y ago|0 comments
13.
Minifying HTML for GPT-4o: Remove all the HTML tags(blancas.io)
149 points by edublancas 2y ago|49 comments
14.
by edublancas 2y ago
there isn't. but you can connect X or LinkedIn. I might add a subscribe button once I get some time :)
15.
by edublancas 2y ago
nothing, I strip out all the HTML tags and pass raw text
16.
by edublancas 2y ago
author here: I'm working on a follow-up post where I benchmark pre-processing techniques (to reduce the token count). Turns out, removing all HTML works well (much cheaper and doesn't impact accuracy). So far, I've only tried
17.
by edublancas 2y ago
author here: I'm working on a follow-up post. Turns out, removing all HTML tags works great and reduces the cost by a huge margin.
18.
Web scraping with GPT-4o: powerful but expensive(blancas.io)
377 points by edublancas 2y ago|167 comments
19.
Using GPT-4o for web scraping(blancas.io)
2 points by edublancas 2y ago|0 comments
20.
Observability for LLM apps with structlog and DuckDB(ploomber.io)
24 points by edublancas 2y ago|10 comments
21.
JSON is all you need: Easily monitor LLM apps with structlog(ploomber.io)
1 points by edublancas 2y ago|0 comments
22.
How good is GPT-4o at generating Flask apps? Surprisingly promising(ploomber.io)
3 points by edublancas 2y ago|0 comments
23.
Deploying VLLM in Production(ploomber.io)
1 points by edublancas 3y ago|0 comments
24.
Rethinking Continuous Integration for Data Science(ploomber.io)
4 points by edublancas 3y ago|0 comments
25.
Effective SQL for Analytics(ploomber.io)
1 points by edublancas 3y ago|0 comments
26.
A systematic overview of prompt engineering(ploomber.io)
2 points by edublancas 3y ago|0 comments
27.
Debugging Streamlit Apps in VSCode(ploomber.io)
2 points by edublancas 3y ago|0 comments
28.
Adding interactive examples to your Python documentation(ploomber.io)
2 points by edublancas 3y ago|0 comments
29.
Interactive Python examples from JavaScript using Jupyter kernels(hidden-truth-8699.ploomberapp.io)
1 points by edublancas 3y ago|0 comments
30.
by edublancas 3y ago
You can get this for pretty much any language by re-using Jupyter kernels; here's a Python example: https://hidden-truth-8699.ploomberapp.io/

More ›