
Research has always shipped answers: an algorithm, an architecture, a paper. 📑 Next, it ships systems that keep getting better after you hand them over. 🎯 My take on RSI that builds RSI. 🔁


Founder & COO @HowOneAI | We deliver production-ready Agentic AI apps that truly work! Founder at EvoAgentX.org | ExLect. @aberdeenuni

Research has always shipped answers: an algorithm, an architecture, a paper. 📑 Next, it ships systems that keep getting better after you hand them over. 🎯 My take on RSI that builds RSI. 🔁


Best plain-English explainer of what Astra's 'recurrent depth' actually is.
ASTRA USES A NEW REASONING TECHNIQUE CALLED "RECURRENT DEPTH". TL;DR: > instead of stacking more Transformer layers, it loops the model's hidden state through the same layer multiple times. some of the reasoning happens internally rather than writing it out as chain-of-thought tokens. why this may help: > more internal reasoning compute without generating a huge CoT > it doesn't keep adding tokens to the KV cache > as @eliebakouch pointed out in his post, recurrent depth isn't necessarily faster. each loop still costs compute and must happen sequentially. the real question is if spending that compute to loop is actually better than investing it on a deeper model. > and there’s a safety concern. CoT reasonings are easily readable. they helped retrace July's Hugging Face attack. so if more reasoning moves into latent spaces, that visibility goes away. > an AI-safety paper co-authored by researchers from OpenAI, Anthropic, Google DeepMind, etc, warned about exactly this last year: “Such latent reasoning models might not need to verbalize any of their thoughts and would thus lose the safety advantages that CoT confers.” > this is exciting because internal reasoning is more similar to how humans think, as we don't speak out everything we think about. > I think it's a matter of time before we see this adopted by more open models, just like how OpenAI’s o1-style reasoning quickly spread across the industry.

Great work
Prime Agent, our new general CLI agent, is out. We show a huge narrative violation where models have the potential to outperform their native harnesses across a variety of evals Huge shoutout to @a1zhang @omouamoua @kevinjosethomas for working on this together

🎮 Can an AI agent keep getting better while it's still running? That's the question behind our next EvoAgentX Talk. We're thrilled to host Seth Karten (PhD @ Princeton, researcher @ Prime Intellect) on Continual Harness: Online Adaptation for Self-Improving Foundation Agents. Instead of fixed prompts and static workflows, Continual Harness lets agents adapt online — accumulating experience, refining their behavior, and reusing what they learn across long-horizon tasks. Seth will show it live through Gemini Plays Pokémon (where harness-level improvements dramatically boost game progress), extend it to ARC-AGI-3, and generalize it to a coding agent that learns from its own past attempts. 🗓️ Thu, Aug 6, 2026 · 10:00 AM (GMT+8) / Wed, Aug 5 · 7:00 PM PDT 💻 Microsoft Teams Join us 👇 details in the comments. #AIAgents #FoundationModels #LLM #AgenticAI #EvoAgentX

What happens when AI scientists can evolve themselves? @_xizhang will share EvoScientist — a system where multi-agent AI collaborates to solve real research problems and continuously improves itself. If you're building AI agents, doing research, or exploring self-evolving systems — this is worth your time. Join us and see how “vibe research” could become reality. 📅 Apr 5, 16:30–17:30 (CN) 🔗 Meeting ID: 470-317-001

I just tested something interesting on HowOne | @HowOneAI There’s a Labubu AI app on the App Store. I uploaded my Labubu toy, and it instantly: - Estimated its value - Generated an animated version from the image That alone was impressive.

Claude Sonnet 4.6 @claudeai is now powering every new App built on HowOne. Smarter reasoning. Cleaner outputs. Stronger vertical logic. But the real unlock? Sonnet 4.6 + Agentic Workflow. One prompt → full-stack AI App → ready to launch. The gap between AI models and AI businesses just got smaller. #howone #vibecoding #BuildingInPublic


It's impossible to learn 100 AI tools, but it is possible to learn one tool that can deliver 100 apps for you.
🚨BREAKING: Non-coders are shipping AI apps right now HowOneAI just silently dropped the 10-minute app builder of 2026. I just shipped a real app in 10 minutes with it: Upload image → 3D miniature transform → cinematic animated video - with no coding Here’s the breakdown: 🧵

Want to build your own personal JARVIS, but find Clawdbot too complex to deploy and understand? Introducing our nanobot — an ultra-lightweight Clawdbot (99% simpler) that gets your personal AI assistant running in under a minute. ⚡️ Delivers core functionality in just ~4,000 lines of Python — 99% smaller than Clawdbot's 400k+ lines 🔗 Fully Open-Sourced: github.com/HKUDS/nanobot Key features of nanobot: - 🪶 Ultra-Lightweight: Just ~4,000 lines of code — 99% smaller than Clawdbot - core functionality. - 🔬 Research-Ready: Clean, readable code that's easy to understand, modify, and extend for research. - ⚡️ Lightning Fast: Minimal footprint means faster startup, lower resource usage, and quicker iterations. - 💎 Easy-to-Use: One-click to depoly and you're ready to go. What can nanobot do for you? - 📈 24/7 Real-Time Market Analysis — Continuous market monitoring and insights. - 🚀 Full-Stack Software Engineer — End-to-end development solutions. - 📅 Smart Daily Routine Manager — Optimize your workflow and productivity. - 📚 Personal Knowledge Assistant — Efficient information organization and retrieval. #Clawdbot #AIAssistant #Agents


Try the app at: k4u37vk-c04fe3.howone.app/

Happy 575th birthday to @UofGlasgow! App made with @HowOneAI. This AI-generated app can turn an image into Line art in seconds.

Don't wait for others to create your app. Describe your desired app and let @HowOneAI make it for you.
Google’s new AI marketing tool Pomelli is on fire lately 🔥 …but here’s the twist: you can build the same kind of platform on @HowOneAI in just 10 minutes ⚡️💡 Try it yourself 👉 apn3vfy-6ff205.howone.app/ #Google #HowOneAI #NoCode #AIagents #MarketingTech #Pomelli

🚀 Launching HowOne (Private Beta) We’re building a self-evolving agentic app builder — turn ideas into fully working agent-powered apps in minutes (users, DB, tools, payments all built-in). Looking for 300 serious beta testers to break things, push limits & shape the product. Perks: ✨ 3 months Pro free ✨ Founding User badge + lifetime 20% off ✨ Early experimental modules ✨ Roadmap voting rights ✨ Featured in Template Marketplace ✨ 1v1 app-design coaching Apply 👉 howone.ai/?invite=PTTAKF #AI #AgenticAI #AITools #AppBuilder #StartupLaunch #IndieDev #Solopreneur #Automation #Makers #AgentWorkflows

Join us and build your own agentic app!
HowOne Beta is officially live. A new way to build real Agentic Apps — no code, no flowcharts, no glue tools. But let me ask you something: Still spending hours learning “vibe coding”? Still stitching together pieces generated by different AI tools just to make one product? Still wiring model keys and configuring APIs for every single workflow? You don’t need to do that anymore. With one prompt, HowOne generates: frontends backends databases multiple agentic workflows…and even self-evolving agents — all deployable in minutes. We’re opening our first wave of beta access. If you want to become an Agentic App Builder, join early and create tools that think, act, and improve — this is your moment. Beta users get: • Free credits • Workflow evolution features • Visual workflow editing • Support to find real use cases & customers • Priority in our upcoming builder marketplace 👇 Join the beta (invite link in comments) The next generation of apps won’t be coded — they will be evolved. #AgenticApps #AIApps #IndieHackers #BuildInPublic #AIStartup