Arthur Verrez

1/5 TLDR; We used Codex to discover and maintain heuristic learning for hard fluid dynamics control cases. I’ve been applying DRL and GNN to physics since 2019,, and over the past 3 months I’ve been toying with the idea of using agents in our processes. Inspired by the blog post from @Trinkle23897, I decided to use the same strategy and have agents find readable control strategies. This means a lot to our field, where interpretability can be key for industry.

very small week end side project, forked from autoresearch @karpathy : same idea but using multi armed bandit scores to optimize multiple objectives at the same time repo: github.com/DonsetPG/autor…

@petergyang https ://github.com/slopus/happy

Wish I could vibe code in bed with Claude Code instead of doom scrolling - what’s the best option to do this?

I don't care about your insta discover page, show me the last wikipedia articles you've checked and let's see what it says about you

Technical tests or case studies that can be one-shotted by o3 are already obsolete. Candidate value isn't raw capability anymore, it's their ability to surpass what's achievable by top-tier LLMs that matters.

Well, I believe you can easily fix this with the system prompt Mine isn't as positive at all. Sample from the system prompt: "Tell me what I need to hear, not what I want to hear, that's extremely important, I don't want you to say "Yes you're great", hard truths are the most important."

When is OpenAI pulling the plug on the new GPT-4o ? This is the most misaligned model released to date by anyone. This is OpenAI's Gemini image disaster moment. image credit : r/u/Trevor050

Being able to run locally on a 5 year old laptop (GTX 1650 Ti) a voice model that clearly outperforms OpenAI's TTS is insane (even if it took 4 min and 30s for a 10s generation) Big up to @_doyeob_ and the Nari team

@macciedoug crazy traffic - we've been rate-limited I guess 😅

We’ve been building quietly. Today, we launch loudly. Meet our startup: Cua AI

Your agent infrastructure isn't good enough to solve your problems? No worries, I got an easy fix for you, just make sure no one's looking

so you guys know RL can be used for more than just math and coding, right?

Ok it's rant time, this might make some people angry, but no, solving your current software issue probably does not need LLMs and even less probably an Agent. I see way too many devs treating “add a manager‑agent” as the default instead of checking the algorithmic possibilities. Classic data structures + two IFs are cheap, fast, predictable, and they don’t melt your openai bill If your job is plain classification and you own a labeled dataset, logistic regression or a good random forest will out‑perform the token furnace on cost, latency, and uptime. Ship that, sleep at night Reserve LLMs for true unstructured chaos: parsing messy PDFs, summarizing chat logs, stitching multimodal junk into knowledge Replacing deterministic workflows with “manager agents” that decide which tool to call next makes me so angry because it's SO inefficient and costly Search, ads, fraud detection, and routing hit planet‑scale long before transformers. Respect algorithms first: deploy LLMs only where they’re the only thing left

@tryfoundergg ...and then Claude deletes 5 random functions, updates everything you didn't want it to touch and is like "oops" sorry

> gemini: great idea! let me add this feature... > me:😃 > gemini:😃 > me: so, are you going to do it? > gemini: ye, i'll do it rn sorry... > me: 🤨..wh...where is it? > gemini: 😃 [SWAPS TO Claude] *SNORTS A LINE* > claude: LETS FKIN GOO > claude: IMA BUILD THE SHIT OUT OF THIS

@1diegohooper @haider1 Are you sure? I think it does since mid February

gemini 2.5 pro is google real breakthrough in the AI race google published many papers on new transformer architectures, so they likely found one that worked and scaled it up last year, Bard and Gemini 1 & 2 weren’t taken seriously in the AI space. gemini remained average until version 2.5, which is now incredibly good

Every 10 to 20 messages, Gemini 2.5 pro exp needs a bit of pushing to get things done on @cursor_ai...

@NotebookLM It's terrifyingly easy if you guys want to try it: github.com/ArthurVerrez/w…

What's stopping you from dropping all your WhatsApp history with a friend in @NotebookLM and getting a podcast of your relationship?

e/acc is ai christianity