Agentic solutions for mission-critical projects. Building https://xitter.cf/t.co/xVI5yzltbB, an agentic OS for professionals, open-source & free. https://xitter.cf/t.co/UFi9A5kXSG is part of @byte5.

Frankfurt, Germany
Joined May 2026
Our Claude meetup on forging skills in Frankfurt was so full last night that the chairs ran out. Every other place to sit was taken too. Thanks for the sharp questions and great conversations. Next one in 3 weeks. Slides (in German): byte5.ai/en/talks
1
1
3
74
An AI agent in a shared team channel can read everything in it. By default, nothing stops it from exposing one person's private context to someone who shouldn't see it. 1/
1
1
49
On the way out, sensitive data is masked before it reaches the model and restored on the way back. The model only ever sees placeholders, never the real values. 3/
1
28
You don't let your AI agent near the real work for 3 reasons: 🚫 It sees data it shouldn't 🚫 It's confident when it's wrong 🚫 When it acts, you can't say who approved it omadia was built to fix all three. Two minutes:
1
1
1
133
Most agents generate answers. omadia delivers results. Eight arguments. One platform. Open source. Free. omadia.ai
1
1
2
293,559
LLMs have 10,000x more working context than your brain. Your brain has 130x more connections than any frontier LLM. Same units. Wild gaps.
🤖 Made with AI
1
1
5
179,366
LLMs guess at math. So our AI agent platform omadia gives agents a headless Excel in memory. Data goes in, the spreadsheet runs the formulas, the model sticks to reasoning, the agent reads the results back. File's there if you ask. Take the guesswork out: omadia.ai
🤖 Made with AI
1
2
1
4
260,225
Your team answers the same questions 20 times a week. 🙄 Build the agent that answers and acts. Right where they work. 💡 ✅ No code. No vendor. No black box. ✅ Self-hosted. Your LLM keys. Your data. GDPR-ready. ✅ Free. Open-source. Made in EU. ➡️ omadia.ai
2
1
2
94,695
2️⃣ Agent memory breaks when every memory is treated as public. omadia.ai uses access-aware graph memory so agents only retrieve memories the current user is allowed to see. That makes shared agents work in real teams, not just private sessions.
2
1
3
133,249
What still annoys you about AI harnesses like OpenClaw, Hermes or PaperClip? What should they do better?
2
32
172,262
Laravel + AI is such a strong combination. Really good to see @taylorotwell putting serious work into making @Laravel a first-class citizen in the emerging AI-native world.
Laravel AI SDK Hackathon at byte5 🚀 Jure introduced the team to the Laravel AI SDK, then we jumped into a hands-on challenge prepared by Hussam: building an AI assistant for a Laravel & Livewire wedding-planning app. Thanks Jure & Hussam for setting up this practical session!
2
42
1⃣ Meet omadia.ai, our new agentic OS: Private data should not become model context. 🚫 omadia tokenises sensitive data before public LLM calls and restores it locally before the answer reaches the user. 💡 The model sees placeholders. The user gets a usable answer. The audit receipt stays PII-free. ✅ Open source, self-hostable. omadia.ai
1
1
3
396
LLMs are not the kind of AI most people think they are. They are a brutally good simulation of it. That ain't a small difference. It's the whole trick.
3
50
91,753
Most AI builders feel like forms with a prompt box. Omadia Builder works more like an engineer. 🛠️ It writes the plugin code, checks whether it actually runs, tests it in a sandbox, and only ships when every check is green. ✅ So you do not just get a generated idea. You get something that has already been checked. ▶️ omadia.ai/builder
3
4
68
446,484