Florin Badita

@Ryanair Why?

no

@outboundman Some of the next steps it’s to give it access to a new server and another domain that I have it in hetzner so both the server and the domain to be controlled via hetzner api and have Claude install Debian 13 and do a full install of everything without human supervision

@baditaflorin going from Debian 13 bare to 70+ self-hosted services in one fork is a bold claim. curious how it holds up in prod

Exciting news, Open Sourcing the AI native server infrastructure that will securely run your whole stack. Fork it, break it, contribute to it. github.com/baditaflorin/S…

@bensig @bensig integrated MemPalace inside ServerClaw github.com/baditaflorin/S…

My friend Milla Jovovich and I spent months creating an AI memory system with Claude. It just posted a perfect score on the standard benchmark - beating every product in the space, free or paid. It's called MemPalace, and it works nothing like anything else out there. Instead of sending your data to a background agent in the cloud, it mines your conversations locally and organizes them into a palace - a structured architecture with wings, halls, and rooms that mirrors how human memory actually works. Here is what that gets you: → Your AI knows who you are before you type a single word - family, projects, preferences, loaded in ~120 tokens → Palace architecture organizes memories by domain and type - not a flat list of facts, a navigable structure → Semantic search across months of conversations finds the answer in position 1 or 2 → AAAK compression fits your entire life context into 120 tokens - 30x lossless compression any LLM reads natively → Contradiction detection catches wrong names, wrong pronouns, wrong ages before you ever see them The benchmarks: 100% recall on LongMemEval — first perfect score ever recorded. 500/500 questions. Every question type at 100%. 92.9% on ConvoMem — more than 2x Mem0's score. 100% on LoCoMo — every multi-hop reasoning category, including temporal inference which stumps most systems. No API key. No cloud. No subscription. One dependency. Runs on your machine. Your memories never leave. MIT License. 100% Open Source. github.com/milla-jovovich…

@cinamarina Built ServerClaw Forkable infrastructure-as-code: bare Debian 13 to 70+ self-hosted services. AI-native. built for Claude Code and AI coding assistants. github.com/baditaflorin/S…

Do you use OpenClaw?

@Shpigford @Shpigford try ServerClaw Forkable infrastructure-as-code: bare Debian 13 to 70+ self-hosted services. AI-native. built for Claude Code and AI coding assistants. github.com/baditaflorin/S…

i turned off my 🦞 the past week of trying to force it to work with other models was a massive drain and i was able to replicate the vast majority of impactful functionality with a mix of claude cowork + code.

@shafu0x Forkable infrastructure-as-code: bare Debian 13 to 70+ self-hosted services. AI-native. built for Claude Code and AI coding assistants. github.com/baditaflorin/S…

WE ARE HIRING send me what you did in the past and what makes you excited about Agentic Commerce DMs are open

Milla Jovovich (actress from The Fifth Element) created a world-beating Claude memory system with @bensig?! - 100% on LongMemEval — first perfect score ever recorded. Free and 100% open source. Github link in the quoted post from Ben. I'm keen to hear how it works for you.

@vasuman No remote work, no interest

No we do not do remote work Yes, we may consider sponsoring a visa for the right candidate

Clearly $5,000 wasn't enough so fuck it - if you refer someone to Varick and we end up hiring them, we will pay you up to $20,000 We are hiring: - Forward Deployed Strategists (referral here is $10k) - Full Stack Engineers (referral here is $15k) - AI Engineers (referral here is $20k) - Tech Leads / Senior Engineers (referral here is $20k) JDs below, please spread this far and wide because we are drowning in contracts and need to move

@OpenAI @OpenAIDevs can we make codex more reliable for power users?

My main bottleneck now it’s not the LLM, it’s my MacBook Pro M1 crashing or freezing when I run 40+ coding’s agents working in parallel in 40 docker repositories.

While I was sleeping yesterday, one of my LLM agent worked for over 5 hours to fully implement a Architecture Decision Record (ADR) Convergence it's in sight.

@Tancrededib Scrapped half the internet, building the intellige layer of the internet domains domainscope.scrapetheworld.org

Investing $250k in non-US based builders. 40 spots, 8 weeks, San Francisco. Flights/housing/food covered + visa support. This Summer Looking for: technical/product builders pre/early idea cofounder-seeking Comment your edge, I’ll get back to you!

@Prince_Canuma Is there a GitHub repository?

Just implemented Google’s TurboQuant in MLX and the results are wild! Needle-in-a-haystack using Qwen3.5-35B-A3B across 8.5K, 32.7K, and 64.2K context lengths: → 6/6 exact match at every quant level → TurboQuant 2.5-bit: 4.9x smaller KV cache → TurboQuant 3.5-bit: 3.8x smaller KV cache The best part: Zero accuracy loss compared to full KV cache.

This morning I made an architectural decision to give my server a voice, the ability to send and receive emails. A few hours later, it used that voice. An email landed in my inbox. From my own infrastructure. No trigger from me. No scheduled job. It just... reached out.

@md_kadiwal domainscope.scrapetheworld.org

Show me what you're building. 👇 No pitch deck. No elevator speech. Just drop the link. I'll click every single one. 🔥

@OmriBuilds AI domains index domainscope.scrapetheworld.org

Pitch your startup in 3 words

@Haezurath Interested, I span many paths

Your personal brand can get you a job. I’m not joking Drop a comment and I’ll give you advice/actionable feedback on how to improve yours

@JesseRank @openhome @menemazarakis domainscope.scrapetheworld.org Building the ai data layer of the internet domains

If you want $100k to work on your project, comment your idea below and I’ll help you land this. I’ll be extra energized to help you land it if you’re building an AI agent that can run on @openhome Shoutout @menemazarakis 👌

@outboundman Sometimes they go în parallel I have around 3 different procente that I handle at a time until one is finished I work on 2 And 3

@baditaflorin the 3-hour batch cycle is basically async software development. design the arch, AI ships 10 ADRs, then you review. completely different rhythm

I'm now working in bursts of designing the architecture and then waiting 3 hours for the AI model to finish, test and then implement another batch of 10 Architecture Decision Records (ADRs) at a time.