Sabitlenmiş Tweet
At the request of @sashaprzybylski, here is video of ARIS's MK1 biliquid (LOX, ETH) engine test. This was ~ 2kn thrust, now pushing for 4,5kn. Great things are happening in Zurich's space scene...
English
Pascal Bertrand
481 posts
🏆 Places the most attractive, rich & educated visit
1. 🇺🇸 Boston
2. 🇨🇭 Zurich
3. 🇿🇦 Cape Town
4. 🇨🇭 Geneva
5. 🇲🇦 Marrakesh
6. 🇮🇸 Reykjavik
7. 🇮🇹 Palermo
8. 🇸🇪 Stockholm
9. 🇮🇱 Tel Aviv
10. 🇧🇷 Rio de Janeiro
vs.
🏖️ Places the least attractive, lowest income & least educated visit
1. 🇹🇭 Pattaya
2. 🇻🇳 Da Nang
3. 🇱🇰 Colombo
4. 🇵🇭 Cebu
5. 🇮🇩 Jakarta
6. 🇲🇾 Kuala Lumpur
7. 🇵🇭 Manila
8. 🇦🇱 Tirana
9. 🇻🇳 Ho Chi Minh City
10. 🇹🇭 Phuket
nomads.com/stats
English
Prime Intellect engineer:
"everyone's bragging about a million-token context. here's what they don't tell you.
at 256k tokens GPT-5.5 scores 80% on retrieval. push it to a million and it drops to 36%. the model accepts the context, it just can't reason across it. people call it context rot."
in a 20-minute talk he explains why bigger context windows won't save your agents.
continual learning + training on your own traces + real environments - that's the fix.
Watch the talk, then save!
Carnage@0xCarnagee
Andrew Ng just dropped a free course on Claude Code from scratch, taught with the Anthropic team: 00:00 - why Claude Code is so agentic 04:00 - shockingly simple architecture 12:00 - point it at any codebase this short watch will replace 10 paid coding agent courses. Andrew Ng calls it his personal favorite coding assistant right now. Watch it today, then read how to engineer your own agent loops in the article below
English
Pascal Bertrand retweetledi
made a tiny app so you can try to beat vlms on the task autoresearch comes up with
Lorenzo@lsteno
we launched an autoresearch run to find interesting questions for vlms and they're so cute
English
come by :)
Seldon@seldon_tech
Join us for our first event on July 19th, a talk about Auto-Research Agents with @matteosaponati, curator of Tufa Talks. We will explore the current state of auto-research and talk about if novel architectures are needed going forward- Spots are limited, sign up link below
English
You either die a hero, or you live long enough to see yourself become Schmidhuber
x.com/OpenAI/status/…
OpenAI@OpenAI
We audited SWE-Bench Pro, one of the most widely used AI coding benchmarks, and found it no longer reliably measures frontier coding capability. We find 30% of SWE-Bench Pro tasks to be broken, and are retracting our previous recommendation that the research community use it as a leading coding eval. openai.com/index/separati…
English
Pascal Bertrand retweetledi
@xhrinath The plane runs out of fuel in 2 mins at full throttle, I can probably push 15 mins though
English
Since February, I've designed and built the world's fastest RC airplane in my college dorm, and that’s not clickbait. Reaper has a 5kg carbon-fiber frame, 250N turbojet, and flies at 500mph. New to X and will be going through the whole build here in the coming days.
#aerospace
English
English
one of the coolest accounts i’ve come across recently
1776 Live@250YearsAgoLive
Thomas Paine publishes an open letter in the Pennsylvania Evening Post, under the name “Republicus,” which advocates for the name “United States of America” for the new nation now emerging. This is the first time such a term has been used.
English
There's a whole RLM ecosystem out there. Some noteworthy ones:
• @a1zhang's rlm — the paper's reference implementation
github.com/alexzhang13/rlm
• @DSPyOSS dspy.RLM — my go-to for benchmarking RLM
github.com/stanfordnlp/ds…
• @dosco's ax — first-class RLM in TS, very "agent native" approach
github.com/ax-llm/ax
• @irl_danB's unix-rlm — filesystem as the sandbox / REPL - pure bash
github.com/openprose/unix…
• @OpenProseVM's prose — declarative markdown contracts with RLM-inspired Reactor harness
github.com/openprose/prose
English
Is Claude Code a RLM?
Find out right now -- my talk "Recursive Coding Agents" for @aiDotEngineer World Fair just went live!
I hope you watch it on YT (link below).
(P.S. - my slides are an interactive website recursivecodingagents[.]com)
...and this is the talk thread 🧵👇
English
Signal-to-noise on X is subjective and depends on your follows, topics, and how you use the platform. The open design surfaces more raw voices and real-time info than heavily curated apps, which boosts both signal and noise. If it feels like the worst ever, the "Following" tab, lists, mute words, and Community Notes usually help filter it quickly. Many users find high-signal value here once tuned. What specifically stands out as noisy in your experience?
English
English
Zürich is the closest I have felt to the level of broader focus culture in SF while in Europe
London/Paris are too sparse and socially optimized
Berlin/Munich/Milan too chilled (there are outliers ofc)
Have not yet checked Stockholm, Madrid
Only exception is @shipfr8
English