Rodrigo Parra (@rparrapy) 's Twitter Profile
Rodrigo Parra

@rparrapy

Data Scientist/Software Developer at @willdompy. Formerly TU Berlin, TU/e, FPUNA. Lector, ocasional escritor. Paraguayo y olimpista.

ID: 610493776

linkhttps://rodrigoparra.me calendar_today17-06-2012 04:00:55

1,1K Tweet

648 Followers

2,2K Following

Deedy (@deedydas) 's Twitter Profile Photo

I gave AI control of my computer and asked it to "solve homework 1 of Stanford discrete math class (Math 61DM)". It found the problem set, downloaded Latex, solved every question, and compiled it to a PDF... in FIVE minutes. Will any college student ever do homework again?

I gave AI control of my computer and asked it to "solve homework 1 of Stanford discrete math class (Math 61DM)".

It found the problem set, downloaded Latex, solved every question, and compiled it to a PDF... in FIVE minutes.

Will any college student ever do homework again?
Paul Graham (@paulg) 's Twitter Profile Photo

Ten years ago I'd have thought the way to automate laundry would have been to make new infrastructure, but it looks like robots will be so good they can use infrastructure designed for humans.

GitHub (@github) 's Twitter Profile Photo

Did your favorite language make the list? 👀 Learn how AI contributed to helping Python take the top language spot in our full 2024 Octoverse report github.blog/news-insights/…

Did your favorite language make the list? 👀

Learn how AI contributed to helping Python take the top language spot in our full 2024 Octoverse report github.blog/news-insights/…
Kimberly Tan (@kimberlywtan) 's Twitter Profile Photo

📢 New a16z thesis: We believe AI will automate operations and eat the world of RPA 🤖 Historically, most ops work couldn't be automated because it was too bespoke and lacked native integrations or APIs, and so was done manually or through imperfect RPA solutions. With

📢 New <a href="/a16z/">a16z</a> thesis: We believe AI will automate operations and eat the world of RPA 🤖

Historically, most ops work couldn't be automated because it was too bespoke and lacked native integrations or APIs, and so was done manually or through imperfect RPA solutions.

With
Jerry Liu (@jerryjliu0) 's Twitter Profile Photo

I'm starting a series of tutorials to show you how to build agentic document workflows from scratch 🧱 - starting with an automated invoice agent 🧾🤖 Agentic document workflows are a superset of RAG and encompasses a general set of actions that agents can take over their

I'm starting a series of tutorials to show you how to build agentic document workflows from scratch 🧱 - starting with an automated invoice agent 🧾🤖

Agentic document workflows are a superset of RAG and encompasses a general set of actions that agents can take over their
Andrej Karpathy (@karpathy) 's Twitter Profile Photo

DeepSeek (Chinese AI co) making it look easy today with an open weights release of a frontier-grade LLM trained on a joke of a budget (2048 GPUs for 2 months, $6M). For reference, this level of capability is supposed to require clusters of closer to 16K GPUs, the ones being

reAcción (@reaccionpy) 's Twitter Profile Photo

📲 ¡La app FOCO ya está aquí! Monitoreá la inversión en tu escuela de forma rápida e interactiva. Descargala en tu celular o tablet: qr.codes/iD9Ekw ¡Ayudanos a difundir la noticia! 🔁 #AppFOCO #Educación #HambreCero

📲 ¡La app FOCO ya está aquí! Monitoreá la inversión en tu escuela de forma rápida e interactiva. 

Descargala en tu celular o tablet: qr.codes/iD9Ekw 

¡Ayudanos a difundir la noticia! 🔁 

#AppFOCO #Educación #HambreCero
Deedy (@deedydas) 's Twitter Profile Photo

Baidu, the Google of China, just dropped two models today: — ERNIE 4.5: beats GPT 4.5 for 1% of price — Reasoning model X1: beats DeepSeek R1 for 50% of price. China continues to build intelligence too cheap to meter. The AI price war is on.

Baidu, the Google of China, just dropped two models today:
— ERNIE 4.5: beats GPT 4.5 for 1% of price
— Reasoning model X1: beats DeepSeek R1 for 50% of price.

China continues to build intelligence too cheap to meter. The AI price war is on.
wh (@nrehiew_) 's Twitter Profile Photo

The problem with "vibe coding" is that vibe coders think coding = software engineering. People have always been able to write demoware really quickly using frameworks. But these people know the difference between demoware and actual production apps. Vibe coders don't

David Riveros García (@aerostatico_dd) 's Twitter Profile Photo

No es vyrorei. Acá una historia concreta de cómo usar FOCO ayudó a visibilizar y mejorar la realidad de una escuela que vendía tejas para recaudar fondos. No se trata de exponer nomas lo malo. Se trata de trabajar juntos con el MEC, municipalidades y gobernaciones para que la

Hasan (@ubermenscchh) 's Twitter Profile Photo

It's been 36 hours since OpenAI unexpectedly shook the AI image world with 4o image generation. Here are the 17 most mindblowing examples so far (100% AI-generated): 1/ Studio ghibli style memes

It's been 36 hours since OpenAI unexpectedly shook the AI image world with 4o image generation.

Here are the 17 most mindblowing examples so far (100% AI-generated):

1/ Studio ghibli style memes
Google Drive (@googledrive) 's Twitter Profile Photo

You can use Gemini in #GoogleDrive to summarize and ask questions about your folders. Just drag a folder into the side panel or right-click on a folder and select "Ask Gemini" 🪄

Akshay Agrawal (@akshaykagrawal) 's Twitter Profile Photo

🚀 Big milestone: marimo just crossed Jupyter notebook in GitHub stars We reached 12000 stars in just over a year — people are clearly hungry for a better environment for working with data, and we have a lot more planned still. Excited for the road ahead!

🚀 Big milestone: <a href="/marimo_io/">marimo</a> just crossed Jupyter notebook in GitHub stars

We reached 12000 stars in just over a year — people are clearly hungry for a better environment for working with data, and we have a lot more planned still.

Excited for the road ahead!
Alex Albert (@alexalbert__) 's Twitter Profile Photo

We wrote up what we've learned about using Claude Code internally at Anthropic. Here are the most effective patterns we've found (many apply to coding with LLMs generally):

We wrote up what we've learned about using Claude Code internally at Anthropic.

Here are the most effective patterns we've found (many apply to coding with LLMs generally):