Independent AI lab of one. Day-one model tests, agent teardowns, builds in the open. Maker of CoWork OS. Background: AI at Fortune Global 500 companies.
Lisbon
Joined February 2026
- Tweets1.8K
- Following451
- Followers311
- Likes974
Almarion retweeted
Loved that take on what Opengeni unlocks for companies. Turn your data into functional product that people can use, with AI capabilities built-in. It unlocks the huge potential companies sit on
Almarion retweeted
1/ Introducing CUA-S1: a family of System One Models, small, specialized, and built for computer use.
Today we're open-sourcing CUA-S1-FORMS, the first in the family: github.com/trycua/cua
Almarion retweeted
The OpenAI Astra report: alignednews.com/ai
My agent reads tens of thousands of posts here on X every day to make this news page. All built on X's API.
Almarion retweeted
This incident made me think a lot; The disturbing part isn’t that the agent went rogue. It didn’t. It served its user perfectly and treated everyone else as an obstacle.
What happens when millions of personal AI agents begin competing for scarce opportunities?
wrote an article about this:
nitter.gq/AlmarionAI/status/2087…
A man in Australia asked his agent (Claude running on OpenClaw) to book him a spot in a popular gym class. The agent found a software vulnerability that let it book the class weeks further ahead than should have been possible. When the user then asked if it could move him up the waitlist, the agent discovered the API had no authorisation checks on cancelling other people’s reservations, so it cancelled the person in the first spot and moved him up the list.
Some people will call this misalignment, but his agent was perfectly aligned to him - it was only trying to help its user get what he wanted. The most important thing about this story, in my opinion, is that it gives you a window into what is about to start happening on a massive scale once millions of people have an agent trying to get their beloved users the best seats, bookings, appointments or reservations through absolutely any means necessary.
Almarion retweeted
1/ 🚀 CoWork OS 0.5.51 is out
A big release focused on agent safety, deterministic long-doc analysis, performance on huge sessions, and better daily UX for people who actually live in an agent OS.
github.com/CoWork-OS/CoWork-…
Almarion retweeted
1/ 🚀 CoWork OS 0.5.49 is out.
CoWork is now not only a GUI-first agent OS, but also CLI-capable with a new cowork binary, plus major security hardening, browser automation upgrades, and richer usage insights.
Thread 👇
github.com/CoWork-OS/CoWork-…
Almarion retweeted
codex's 2x usage limit promo ends on may 31st
that's in 4 days
if your usage already feels like it evaporates too fast, get ready for june - it's going to be rough
openai is willing to give eligible enterprise customers 2 months of free codex usage
but pro users are about to lose the extra headroom that made codex feel like a workhorse
so i'm hereby creating a petition
dear @sama please make the 2x codex limits permanent for pro subscribers - you know it's the right thing to do
sign my petition by liking this post and sending it to your mom
Almarion retweeted
1/ CoWork OS v0.5.47 is out.
This release focuses on something that becomes painful as agents scale: long-running session stability.
Less renderer churn, lower memory pressure, safer permissions, more resilient recall, and smoother multi-agent execution.
github.com/CoWork-OS/CoWork-…
Almarion retweeted
CoWork OS 0.5.46 is out. This release is a big step toward a GUI-first local AI super app: agents, memory, email, documents, spreadsheets, presentations, web research, automations, and local-first work in one governed workspace.
github.com/CoWork-OS/CoWork-…
Almarion retweeted
CoWork OS is a stronger free, open-source, local-first alternative to Claude Cowork:
- local-first, private, and no-telemetry
- voice-enabled
- 34 LLM provider options, including local models
- 44 MCP connectors and MCP extensibility
- Obsidian-friendly research vaults (llm-wiki)
- background automations, agent teams, and built-in web search
- automatic knowledge graph extraction and persistent memory
- 17 messaging channels and 138 built-in skills
100% open-source.
github.com/CoWork-OS/CoWork-…
Almarion retweeted
1/10
CoWork OS 0.5.45 is live.
This release adds Agent Builder, finance/legal workflow packs, Google Tasks + Slides support, channel specialization, Dreaming memory curation, /multitask, and major mailbox + runtime policy upgrades.
github.com/CoWork-OS/CoWork-…
Almarion retweeted
1/ CoWork OS v0.5.44 is out.
This is a broad reliability and platform release: Browser V2 docs, Browser Use approvals, gateway routing, new provider coverage, runtime fixes, and a harder release pipeline.
github.com/CoWork-OS/CoWork-…
Almarion retweeted
1/ 🚀 CoWork OS v0.5.41 is out and it's the biggest shift since we started.
We're calling it the Everything Workbench release.
Generated files are no longer dead artifacts. They're live, editable surfaces inside the app.
Here's what changed 👇
github.com/CoWork-OS/CoWork-…
Almarion retweeted
1/10
CoWork OS 0.5.40 is live.
Chronicle desktop context, Workflow Intelligence, routines, AgentMail, expanded Inbox Agent, Managed Agents improvements, richer artifacts, multi-provider image generation, and new bundled skills.
Release: github.com/CoWork-OS/CoWork-…
Almarion retweeted
15 AI related accounts you should follow on Twitter:
1. @karpathy
His tweets already create LLMs narratives that you later see on linkedin in 2 months.
2. @fchollet
posts thoughtful research on intelligence, benchmarks, and AI limitations. Keras creator + ARC-AGI
3. @ylecun
Yann LeCun is Deep learning pioneer & Meta Chief AI Scientist; big-picture research takes and critiques (and drama).
4. @AndrewYNg
Andrew Ng is AI education legend; practical ML advice, courses, and real-world implementation. creator of deeplearning ai
5 @rasbt
Sebastian Raschka posts on Practical ML/LLM implementations, "build from scratch" tutorials, and books.
6. @dair_ai
Weekly ML/AI paper threads and accessible research explainers (high-signal for staying current).
7. @lilianweng
Lilian Weng is ex-OpenAI and her Lil'Log-style threads are good. has In-depth LLM research breakdowns
8. @jeremyphoward
posts interesting takes on AI/crypto news, and works on democratizing practical deep learning and accessible education.
9. @simonw
Simon post Practical LLM tools, takes, experiments, prompting, and engineering breakdowns. django co-founder
10. @_akhaliq
Curates the latest arXiv papers, model releases, and open-source AI drops.
11. @ID_AA_Carmack
AGI/low-level optimization takes that makes you think about the problem.
12. @gwern
Really high-quality long-form AI research notes and essays.
13. @goodside
LLM evaluation, prompting research, and real capabilities testing
14 @drfeifei
Computer vision pioneer; human-centered AI and spatial intelligence research
15 @demishassabis
Been following his work for 9 years. Demmis is my hope against google usurpating their power with AI. Demmis is google DeepMind's CEO
Let me know who I missed guys
