ClawCamp Campfire: Sovereign AI - Own Your Stack

Join this special ClawCamp Campfire to dive deeper into Sovereign AI, and owning more of your agentic AI Stack. In a fast evolving agentic world where privacy is becoming harder and harder to maintain, and questions continue to rise around where your data goes and how it's being used by service providers; sovereign AI models, harnesses, and tools are evolving alongside their frontier counterparts. Join us to explore Sovereign AI options that allow you to own more of your agentic AI Stack, in order to maintain privacy and control access to your intellectual property. This Campfire is hosted at the Luxor AI Cloud office, home of Tenki, and every attendee gets $500 in Tenki credits to run their own sandboxes (details below). Learn more at ClawCamp.org

Want to Mentor, Speak, Sponsor, Volunteer, or Partner? Fill out our contact form: Aiify.io/form

📅 Schedule 6:00pm: Doors Open 6:30pm: Your Editor Local, Your Compute Anywhere: Crabbox + Tenki Sandboxes by Colin Lowenberg 6:45pm: Local AI Workshop by Jason Stine 7:40pm: Big Models, Little Machines: Distributed AI Workshop by Kirah Sapong 8:15pm: Scoping and Permissioning AI for a Shared Future by Rayyan Zahid 8:30pm: Networking and Wrap up 9:00pm: Doors Close 🎁 $500 in Tenki credits for every attendee Tenki by Luxor.ai is sponsoring this Campfire with $500 in cloud credits per attendee, no credit card required. Spend them on Sandboxes (Linux VMs your agent can run code in), GitHub Actions Runners, or Code Reviewer. The fastest way in is to let your agent set itself up. Paste this into OpenClaw, Claude Code, Codex, or whichever harness you use: Onboard me to Tenki. Follow the agent instructions at https://tenki.cloud/skill.md and set me up end to end.

📦 Your Editor Local, Your Compute Anywhere: Crabbox + Tenki Sandboxes by Colin Lowenberg Running a model on your laptop keeps your prompts private. Running your agent's code on your laptop is a different problem: an agent that can run shell commands can also delete your home directory, and a laptop can only hold so many parallel jobs. Crabbox (https://crabbox.sh), the new CLI from the OpenClaw team, solves this by keeping your editor and working tree local while running every command in the box you choose: a local container, your own SSH host, or a managed sandbox. Tenki is one of its providers, so the same crabbox run you use on your laptop can run on a fresh Linux VM in the cloud and stream the result back. In 15 minutes, Colin will show: crabbox run --provider tenki -- pnpm test from a local checkout, with the sandbox created, synced, run and torn down for you. The OpenClaw Tenki sandbox backend (https://github.com/LuxorLabs/tenki-openclaw), so every tool call your OpenClaw agent makes runs in a remote VM instead of on your machine. An OpenClaw gateway running entirely on a Tenki sandbox. Live demo at https://claw.moi Worked examples to follow along with or try later: Crabbox Tenki provider docs: https://crabbox.sh/providers/tenki.html Crabbox + Tenki cookbook: https://github.com/LuxorLabs/cookbook/tree/main/examples/crabbox-tenki OpenClaw Tenki plugin: https://github.com/LuxorLabs/tenki-openclaw

💻 Local AI Workshop by Jason Stine Ever feel uncomfortable about sharing too much personal info with ChatGPT/Gemini/Claude but still want AI’s help? Against giving Big Tech your private life on a golden platter running on wasteful resource-sucking data centers? This workshop is for you! Earlier this year there was a breakthrough - we can now run AI that are more than capable of being private and even completely offline assistants on personal computers and laptops! This is also a workshop for anyone who find setting up agents with OpenClaw too intimidating but still want a capable AI experience like ChatGPT that keeps all private information on their own computers - without writing any code or running any terminal/command line commands! What we'll cover: Setting up Unsloth Desktop (open source ChatGPT-like interface) and running powerful open source AI models of your choosing on your own computer. Learning important concepts about running AI on your own computer - from why open source AI exists to tokens, context, model quantization, MCP tool servers, and more. This is a hands-on workshop - bring your computer and get ready to download and install some AI! System Requirements: The more RAM your computer has, the better. It’s now possible to run near-frontier-capable-AI on computers with just 16GB of RAM. That said, even if your computer has less than that (like 8GB), you can run less powerful AI that is still impressive and are welcome to attend! To be efficient, we encourage participants to download and install Unsloth Desktop (https://unsloth.ai/docs/desktop) as well as let it automatically download a suitable AI model it selects for you (if any) when first running the app in advance before coming to the event. During the event, you’ll learn how to find even better AI models to add to your Unsloth Desktop setup.

🛠️Big Models, Little Machines: Distributed AI Workshop by Kirah Sapong Running AI models on your own computer is exciting, but even powerful laptops have limits. By connecting multiple computers together, we can combine their resources to run larger AI models than any one device could handle alone. In this workshop for all levels, we’ll explore distributed AI inference: how multiple computers can work together to run AI. We’ll demonstrate how to split AI models across devices using Aquaduck, without writing any code! What we’ll cover: Understanding how AI models run on your computer and why memory matters. ⁠Learning how multiple computers can work together to run AI. Exploring model splitting and how it enables larger AI models to run across devices. Getting hands-on with Aquaduck to connect computers and run AI together. Bring your laptop and get ready to experiment with running AI across multiple devices. You can download and install Aquaduck (https://aquaduck.ai) ahead of time so you can jump straight into the hands-on activities. No prior AI experience or programming knowledge is required.

✅ Scoping and Permissioning in the Age of Personal AI by Rayyan Zahid Soon everyone will have a personal agent that knows their work, their family and their friends, and that agent will talk every day to other people's agents and to public ones. For that to work, each agent has to know what it may share, who it speaks for, and how far its permissions reach when it passes work along. Rayyan talks about what it takes to build multi-agent systems that people can trust with their whole lives.

💡 Sponsors & Partners

🎤 Speakers & Mentors Colin Lowenberg: Colin works on Tenki, the cloud sandbox platform from Luxor.ai, and runs Dabl Club, a builder community in San Francisco. He spends most days putting coding agents on remote VMs and keeping them off his laptop. Jason Stine: Jason is actually in no way related to Unsloth. He's just a passionate AI hobbyist who enjoys mentoring others and found that Unsloth Desktop was the most straightforward software for this workshop. He's a fullstack web developer (software engineer) who uses Cursor AI every day at work. Kirah Sapong: Kirah is a co-founder at Supermodel Labs, working on AI systems. Previously, she worked on model evaluations and AI platform products at Together AI, decentralized systems and developer platforms at Block, and is currently pursuing an M.S. in Artificial Intelligence at the University of Pennsylvania. Her research interests include agent systems, inference optimization, and AI economics. Rayyan Zahid is a builder in San Francisco, working in stealth on planetary scale AI. He facilitates Vibe Coding Nights, founded Immersive Commons (a 300+ member builder community at Frontier Tower), is a Nebius Fellow, and advises Cotal, an open source space where AI agents hand work to each other.

🦞 About ClawCamp ClawCamp.org is a non-profit organization dedicated to advancing personal AI proficiency. Events bring together organizations, experts, and mentors to provide the tools, training, and support people need to advance. Beginners acquire powerful skills to achieve almost any goal, while intermediate and advanced attendees dive deeper into more nuanced aspects of agents, harnesses, LLMs, and related technologies that drive AI’s potential even further. ClawCamp’s community of AI enthusiasts support one another in advancing their AI mastery and celebrate each others’ successes, big or small. Visit ClawCamp.org for the latest updates.

🌐 Connect with ClawCamp Luma Calendar:  lu.ma/ClawCamp Discord: discord.gg/Yc5hWZUWuC Linkedin:  linkedin.com/company/clawcamp Facebook: facebook.com/clawcamp Instagram: instagram.com/clawcamp X/Twitter: x.com/ClawCampus

💻 About Aiify.io The SF Chapter of ClawCamp.org is organized by Aiify.io : Aiify.io hosts events, workshops, hackathons, accelerators, and more, for developers, integrators, businesses, designers, scientists, clinicians, and other professionals to explore how AI can be leveraged for practical solutions and increased efficiency – today and into the future. Follow and engage with us here: Luma Calendar: lu.ma/Aiifyio Linkedin: linkedin.com/company/aiify-io Telegram: t.me/aiify_io X/Twitter: x.com/aiifyio Discord: discord.gg/Yc5hWZUWuC *By registering for this event, you agree to release all organizers and their affiliates from all liability related to your registration and attendance. You also consent to being added to organizer email list(s) for updates and notification, including future events and resources. You understand that you may unsubscribe from any email lists at any time. You also understand that media may be recorded at this event and you agree to hold the event organizers and their affiliates harmless in relation to any incidental inclusion of your likeness in shared media.