TikuOS:Simple. Ubiquitous. Intelligence,Everywhere

An open-source operating system for physical AI endpoints. Deploy-and-forget devices that run for years on a coin cell or indefinitely on harvested energy.

[Pablo T. @controlpaths]

A 35B parameters model (with 3B active per token) in an embedded board. Next months are going to be insane.

[Rohan Joshi @ron_joshi]

Introducing Kitten TTS V0.8: open-source TTS that fits in 25MB.

Three variants: 80M | 40M | 14M (<25MB)

Highly expressive. Runs on CPU. Built for edge.

No GPU? No problem. Ship voice anywhere.

Check it out:

[Daniel Liu @p1nosaur]

chat with your kicad components, answers directly from datasheet

[Vitto Rivabella @VittoStack]

I’m building a mini robot for my OpenClaw.

- 4 motors
- WiFi and Bluetooth integrated
- 3 hours of autonomy
- ESP32 powered

Only 7cm x 6cm.

[Nav Toor @heynavtoor]

🚨 BREAKING: Someone just rebuilt the entire AI assistant stack in Zig.

It's called NullClaw. The binary is 678 KB. It uses ~1 MB of RAM. It boots in under 2 milliseconds.

No runtime. No VM. No framework. No garbage collector. Just raw Zig.

Here's why this is absurd:

→ OpenClaw needs a $599 Mac Mini and 1 GB+ RAM
→ NanoBot needs 100 MB+ RAM and Python
→ PicoClaw needs 10 MB RAM and Go

NullClaw runs on a $5 board with 1 MB of RAM.

Same functionality. 0.1% of the resources.

Here's what's packed into that 678 KB:

→ 22+ AI providers (OpenAI, Anthropic, Ollama, DeepSeek, Groq, etc.)
→ 13 chat channels (Telegram, Discord, Slack, WhatsApp, iMessage, IRC)
→ 18+ built-in tools
→ Hybrid vector + keyword memory search
→ Multi-layer sandboxing (Landlock, Firejail, Docker)
→ Hardware peripheral support (Arduino, Raspberry Pi, STM32)
→ MCP, subagents, streaming, voice, the full stack

Here's the wildest part:

Every subsystem is a vtable interface. Swap any provider, channel, tool, memory backend, or runtime with a config change. Zero code changes.

It even encrypts your API keys with ChaCha20-Poly1305 by default.

2,738 tests. ~45,000 lines of Zig. Zero dependencies beyond libc.

100% Open Source. MIT License.

[Ilir Aliu @IlirAliu_]

A highly affordable, fully 3D-printed robotic arm.
[📍 open source]

Features:
> Fully 3D printed design
> 7+1 DOF
> or physical AI research & imitation learning

Zero custom hardware needed, just print and add your servo kit!

> URDF files
> STEP & STL files
> Step-by-step assembly manual
> Ready-to-print Bambu Lab projects

Great project, @batsuev_es!

Download and build yours here:

https://t.co/azq8cbBYzS

——

Weekly robotics and AI insights.
Subscribe free: https://t.co/9Nm01QUcw3

[i2cjak @i2cjak]

an actually USEFUL MCP that is ALREADY HOOKED UP TO MOUSER/DIGIK*Y/JLCPCB???

you've gotta try it out anon. i used this guy's previous iteration that was JLC only, it's good. It's really good.

[Massimo @Rainmaker1973]

Japanese HAGAMOSphere, a drone that rolls and flies.

[Robert Scoble @Scobleizer]

What was I talking about yesterday?

OpenAI can put @openclaw on a small device.

I could see buying several and leaving one in each of our cars.

[Ejaaz @cryptopunk7213]

i fucking love this.

dude slapped an OpenClaw agent on a rasperry pi + mic + led screen

result? a palm-sized ai agent in your pocket that you can speak to and it just goes away and does stuff!

a tiny device that costs <$100 but does the work of a 6-fig employee

i desperately want to see more homeware AI devices in people’s hands. this is fucking cool.

[Louis Gleeson @aigleeson]

This is wild 🤯

A developer on GitHub just shipped a complete AI agent that:

• Runs on 256MB RAM
• Has 11 built-in tools
• Keeps persistent memory
• Ships with Telegram integration

And it’s a single 12MB file.

We’ve been overcomplicating agents.

Link + description 👇

[Tech with Mak @techNmak]

OpenClaw's success sparked an explosion of alternatives.

6 different implementations. 6 different philosophies. Same core inspiration.

Here's the breakdown:

𝗡𝗮𝗻𝗼𝗯𝗼𝘁 (Python)
→ ~4,000 lines of code (99% smaller than OpenClaw)
→ Research-ready, clean, readable
→ MCP support, multi-channel
→ "Ultra-lightweight personal AI assistant"

𝗡𝗮𝗻𝗼𝗖𝗹𝗮𝘄 (TypeScript)
→ "Small enough to understand in 8 minutes"
→ Agents run in actual Linux containers
→ First to support Agent Swarms
→ Philosophy: Fork it, customize it, own it

𝗜𝗿𝗼𝗻𝗖𝗹𝗮𝘄 (Rust)
→ Security-first design
→ WASM sandbox for untrusted tools
→ Credential protection, prompt injection defense
→ "Your AI assistant should work for you, not against you"

𝗭𝗲𝗿𝗼𝗖𝗹𝗮𝘄 (Rust)
→ Runs on $10 hardware with <5MB RAM
→ <10ms startup time
→ Trait-driven architecture, swap anything
→ "Zero overhead. Zero compromise."

𝗣𝗶𝗰𝗼𝗖𝗹𝗮𝘄 (Go)
→ <10MB RAM, 1s boot
→ Runs on old Android phones
→ 95% AI-generated codebase
→ Ultra-efficient, runs on any Linux board

𝗧𝗶𝗻𝘆𝗖𝗹𝗮𝘄 (TypeScript)
→ Multi-agent, multi-team, multi-channel
→ Team collaboration with chain execution
→ Live TUI dashboard for monitoring
→ "24/7 AI assistant"

OpenClaw proved the demand. These projects are exploring different trade-offs.

The future of AI assistants is open source, forkable, and runs on anything.

[sunny madra @sundeep]

Edge devices like this will become thin clients that leverage massive amounts of intelligence in the cloud

[Vitto Rivabella @VittoStack]

I’m giving my OpenClaw a body.

Update #2:

- Made a cardboard prototype
- Can edit its own code (taught itself to rotate 360°)
- Added vision + basic obstacle detection
- Shipped “Follow” and “Seek” modes (tracks people, searches objects)

It’s getting… autonomous

[Marco Franzon @mfranz_on]

Finally I can chat with my ESP32.

Thanks to zclaw project you can interact with the board with a set of tools LLM enriched.

You can get sensor info, change status simply using natural language.

Pivoting into hardware will be so funny now.

[TikuOS:Simple. Ubiquitous. Intelligence,Everywhere @Bfreecomputers]

@madietlx Awesome 🙌

[Sebastian Völkl @basti_vkl]

built my own personal assistent device that runs OpenClaw.

I was curious what the smallest form factor could be that fits in my pocket so I wanted to use the Pi Zero W.

Works via Push to Talk-&gt;Transcribe-&gt;Sends to OpenClaw and streams the response back.

[Natalie @livinoffwater]

This is actually huge

[Brian Roemmele @BrianRoemmele]

https://t.co/ErCFfVGmjT

[TikuOS:Simple. Ubiquitous. Intelligence,Everywhere @Bfreecomputers]

Send useful.

[Tom Dörr @tom_doerr]

Framework for quadruped robots on Arduino and Raspberry Pi

https://t.co/MSFrTKHwVz