Guillaume Loret

🇪🇺 in the 🇺🇸

(background picture by Nicolas Raymond:

[The New Stack @thenewstack]

Why GPT-5.4, Claude, and Gemini can't agree on basic, real-world facts https://t.co/2Vc6eh3UK2

[Android Developers @AndroidDev]

Android Bench evaluates LLM coding performance by focusing on the nuances and platform expertise that are unique to Android.

See how your favorite model stacks up → https://t.co/Ckv3kG4soG

Are the results what you expected? 👀 Let us know in the replies!

[Guillaume Loret @yom]

Open source is facing a double crisis: we’re relying on "zombie" packages with no human maintainers while using flawed metrics that ignore our most critical infrastructure.

* https://t.co/k4DAIh6djo
*
https://t.co/EDcT6nE80V

[Simon Kuestenmacher @simongerman600]

Wait, there is also an English version: https://t.co/SCbz8HjZ6h

[Guillaume Loret @yom]

Researchers are finding that subtle inconsistencies in physical laws, like light reflections, shadows, and perspective lines, are still reliable indicators for identifying deepfakes.

https://t.co/3KOYykJTHx

[Guillaume Loret @yom]

Using a hybrid architecture for AI agents: files as the human-readable "interface layer" for context, while leveraging databases as the scalable "storage layer" for long-term recall.

https://t.co/BK8W5yo2Vv

[Guillaume Loret @yom]

Despite its power, WebAssembly often requires complex JavaScript "glue code" just to talk to the browser. The article explores how the WebAssembly future could look like.

https://t.co/JENpet8Xgo

[Guillaume Loret @yom]

https://t.co/IUiPOAeeFY

[Guillaume Loret @yom]

Contrary to the promise of reduced workloads, new research shows that AI often intensifies work. Because the tech makes "doing more" feel accessible, employees are expanding their job scopes, multitasking more, and blurring the lines between work and downtime.

[Guillaume Loret @yom]

Query knowledge graphs without the hassle: GRASP lets you run complex SPARQL queries with simple natural language questions

https://t.co/zVm8axaUJZ

[Guillaume Loret @yom]

https://t.co/FHkdyrUsNC

[Guillaume Loret @yom]

Tech debt isn't always bad! Like financial debt, it depends on the interest rate. High-interest tech debt can cripple your team, but low-interest tech debt can help you move faster. Check out this great read by Erik Schluntz on managing tech debt effectively.

[Guillaume Loret @yom]

https://t.co/CiZYsMAAKI

[Guillaume Loret @yom]

LangChain is moving away from Retrieval Augmented Generation (RAG) and towards agentic search.
[https://t.co/x0jvCNiERC

[Guillaume Loret @yom]

https://t.co/I40rtaauyl

[Gilles Babinet @babgi]

Ce visuel de Bloomberg est excellent (et un peu terrifiant aussi)...
C'est l'explication de ce qu'est une bulle et là il s'agit d'une bulle dans l'IA. Quand les investissements deviennent circulaires, sans irriguer l'économie réelle
- OpenAI achète des puces Nvidia
- Nvidia investit dans OpenAI
- Oracle loue des puces Nvidia à OpenAI
- OpenAI prend des participations dans AMD
Fasten your seatbealts, it gonna be shaky.
On rappellera que la somme totale des engagements pris par OpenAI se chiffrent à 1000 milliards de dollars.

[Guillaume Loret @yom]

His journey in AI: finding clients beats building. AI accelerates, niches vanish, judgment & personal brand are the real moats. Be realistic.

https://t.co/MdPaFlhAEv

[Omer Cheema @OmerCheeema]

Someone at a16z claimed a few weeks ago that 80% of Bay Area startups are building on Chinese open source models. The graphic below shows Chinese model downloads exceeding US models on HuggingFace.

[Deedy @deedydas]

60% of CS PhDs in the US are not American.
53% of CS Masters grads in the US are not American.
BigTech cos have <15% employees on H-1B.

BigTech hiring isn't necessarily biased against Americans; most Americans just don't study advanced CS!

[Guillaume Loret @yom]

https://t.co/JbGrKSlmB4

[Guillaume Loret @yom]

For over 170 years, as cities evolve from craftsmanship to technology, they maintain a consistent "economic coherence."

It measures how well the mix of jobs, industries, and technologies in a city fit together.

[Guillaume Loret @yom]

Constrained Transformation: Cities can develop new industries and abandon old ones, but they must do so while maintaining coherence. Even cities like Pittsburgh and Boston that shifted from heavy manufacturing to high-tech had to find paths that preserved their economic coherence