Pedro Madruga
banner
pedromadruga.com
Pedro Madruga
@pedromadruga.com
Applied AI Scientist (in 🇩🇰 since 2012) • http://pedromadruga.com • Interested in Information Retrieval at scale • Lead AI Scientist at Karnov Group

Opinions are my own.
This year, I've tried Helium and Waterfox browsers. I found the former to be buggy and there's something about Chromium-based browsers that doesn't bode well with me. And I've always been a Firefox user, since its Firebird days.

Waterfox was perfect since the get-go. Both on Mac, Linux and Android.
December 23, 2025 at 9:18 AM
This was also a year where I have been migrating to European-based services. Went from iCloud to Proton Drive, Gmail to Proton Mail and so on.

Migrated from iPhone to Android, since I bought a FairPhone and am very happy with it. FP is much better than I anticipated!
December 22, 2025 at 7:18 PM
Doing a "year in review" on the technologies I've tried/read about.

One thing that comes to mind is the hostility towards MCP, especially by people who didn't try it.

And I used to do Javascript, which got a lot of heat back in the day. MCP is just a Protocol, people.
December 21, 2025 at 7:45 AM
I just read "AI Agents in Action". It's well-supported by code and drawings. Being quite heavy on OpenAI and Microsoft tooling is a drawback. Online reviews are pretty mixed as well.

It's not a beginner-level book but needs more depth. Decent read, all in all.

www.manning.com/books/ai-age...
AI Agents in Action - Micheal Lanham
Create LLM-powered autonomous agents and intelligent assistants tailored to your business and personal needs.
www.manning.com
December 20, 2025 at 10:13 AM
Reposted by Pedro Madruga
📢 Launching EuroLLM-22B - a fully open, EU-made LLM supporting all 24 EU languages (+11 more)!

🤝 Built by a Europe-wide consortium with #HorizonEurope & @eurohpc-ju.bsky.social support.

🔓 Available now on Hugging Face: lnkd.in/e-xgQXTX

🔗 Learn more: lnkd.in/e6FdpYrN
December 15, 2025 at 1:11 PM
Reposted by Pedro Madruga
Dolphin-v2 🐬 new document parsing model released by Bytedance

huggingface.co/ByteDance/Do...

✨ 3B - MIT license
✨ Works on any document: PDFs, scans, photos
✨ Understands 21 types of content: text, tables, code, formulas, figures & more
✨ Pixel-level precision via absolute coordinate prediction
ByteDance/Dolphin-v2 · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
December 12, 2025 at 12:54 PM
The influence of domain knowledge is huge when it comes to applied AI.

Domain expertise can prevent overengineering and/or unnecessary costs.

Working in the legal industry as an AI scientist it became obvious early on how crucial it is to build the bridge towards domain experts.

Every day.
December 10, 2025 at 10:13 AM
A chatbot that interacts with NeurIPS 2025 papers: neurips.zeroentropy.dev

Went in to find Information Retrieval related papers and there's a few. Happy reading!
December 9, 2025 at 10:13 AM
Reposted by Pedro Madruga
We tested one of the most common prompting techniques: giving the AI a persona to make it more accurate

We found that telling the AI "you are a great physicist" doesn't make it significantly more accurate at answering physics questions, nor does "you are a lawyer" make it worse.
December 8, 2025 at 6:32 PM
We have come full circle.
December 8, 2025 at 8:45 AM
Reposted by Pedro Madruga
TITANS & MIRAS: real continual learning

MIRAS = a unifying theory of transformers (attention) and state space models (SSM, e.g. Mamba, RNNs)

TITANS = an optimal MIRAS implementation that’s “halfway between” SSM & transformer with a CL memory module

let’s dive in!

research.google/blog/titans-...
Titans + MIRAS: Helping AI have long-term memory
research.google
December 6, 2025 at 4:04 PM
I am a fan of SDD but it often gets confused with Vibe Coding, specially in companies where innovation is met with resistance.

This articles shares some criticism, yet by someone who actually tried it. I disagree with the article and it seems that HN people in thread do too.

Give SDD a try!
Spec-Driven Development: The Waterfall Strikes Back | Hacker News
Its not a surprise to me that this approach also helps AI coding agents to work more effectively, as in-depth planning is essentially moving the thinking upfront.(I wrote more about this here:…
news.ycombinator.com
November 18, 2025 at 11:01 AM
I have been trying lefthook and damn this thing is fast. Also, the ability to run scripts is handy if you like to keep the config tidy.
GitHub - evilmartians/lefthook: Fast and powerful Git hooks manager for any type of projects.
Fast and powerful Git hooks manager for any type of projects. - evilmartians/lefthook
github.com
November 17, 2025 at 6:08 PM
“Before being technical, science is visionary”
Carlo Rovelli, Reality Is Not What It Seems
November 15, 2025 at 9:28 AM
Business-applied AI development requires a much higher level of experimentation than traditional software development.

This is because AI development requires a very tight connection to domain knowledge. And each domain requires its own approach, for the most part.
November 14, 2025 at 10:00 AM
A short (way too short) yet fantastic reading. Rovelli has the gift of explaining complex things simply.

I have quickly moved to reading The Order of Time, another of his books. The way he explains time is just so accessible - thus mind blowing.
November 11, 2025 at 6:45 AM
It’s reasonably amusing to see the resistance from some traditional software developers to some AI tooling. MCP for example, gets a lot of focus from tech doomers because it’s new.

It’s Ok to not use those don’t fit a use case but until they try it it’s just preemptive criticism.
November 10, 2025 at 11:30 AM
TIL!
I started using the ThunderAI add-on in Thunderbird. Now my local LLM automatically classifies, auto-replies (reviewable), and summarizes emails across multiple accounts. I did not expect to appreciate it this much!

#Thunderbird #ThunderAI #Productivity #PrivacyFirst #LLM
November 7, 2025 at 8:13 PM
The term Agentic Coding is something that resonates quite a lot. But I’ve heard Spec Driven Development and it is spot on.

Both these are almost polar opposite to what Vibe Coding is (imo), even though all are AI powered.

Also, for prototyping: vibe coding all the way.
November 4, 2025 at 1:33 PM
An important component of the success of AI-based products is the ability to intertwine data science and (software) engineering.

Enforcing a separation of those is a recipe for failure.

When it works though, it feels like a dance: there might be toe-stepping but the song is the same.
September 12, 2025 at 7:54 PM
When people ask what am I doing with so many Pi’s
September 7, 2025 at 6:06 PM
Reposted by Pedro Madruga
this seems like a very good idea actually
August 15, 2025 at 1:38 PM
Phi4 didn’t even try
August 12, 2025 at 7:16 PM