Blog
Insights on agentic development, multi-agent workflows, and building developer tools.

Designing Mobile Workspaces for Agentic Coding
Why typing code on glass feels frustrating, but orchestrating AI agents on an iPhone or iPad feels natural — and how we built an interface for real mobile engineering workflows.
Read moreAll Articles
- News
MadoHub for iOS 1.1.2: Streaming Visibility, Compact Controls, and Accessibility
Smoother long-reply streaming, one-tap mid-stream cancellation, a more ergonomic compact composer, and refined VoiceOver navigation.

- News
MadoHub for iPhone and iPad Is on the App Store
A visual AI workspace for iPhone and iPad: give each Agent a role, arrange them on a canvas, and connect them into a workflow you can follow.

- News
MadoHub Companion: Mirror Your Canvas to Your iPhone
A new iOS app pairs with your Mac in seconds and mirrors your canvas over the local network — no cloud relay, no account, your conversation stays on your Wi-Fi.

Claude Code's Completion Signals Keep Changing. Here's How We Keep Up
Why keyword-matching on Claude Code's completion banner kept breaking, and what MadoHub actually trusts to know when an agent's turn is over.

Local, Private Voice Input with On-Device Whisper
Why MadoHub runs speech recognition on your machine instead of calling a cloud API, what that means for privacy and cost, and how to get the most out of dictation.

- News
Day-One Support for Claude Code Dynamic Workflows
When Claude Code fans a session out into many subagents, MadoHub keeps the canvas readable — one progress line per tile, no badge flooding, with token usage rolled into the Context Gauge.

- News
One Drawer for Flows, Exploration, and Your Team
Flows, Exploration, and Team Hub now share a single tabbed rail on the right edge — and Team Hub got denser with state dots, a task board, and a message tail.

- News
Introducing MadoAgent: Orchestrate Your Canvas in Plain English
A persistent panel that turns 'spawn an agent, wire it up, hand it a task' into a sentence instead of a sequence of clicks — opened with Cmd+Shift+M.

- News
Orchestrator Mode: Fire-and-Forget Codex Workers
Ask MadoAgent to hand a self-contained task to a headless Codex worker — no terminal tile to babysit, with reliable completion detection and sandboxed by default.

Teaching MadoHub to Remember
Inside MadoHub's memory system: six kinds of memory, namespacing by persona and project, and why nothing an agent learns becomes permanent without a human looking at it first.

Mission Control: Never Lose Track of Which Agent Needs You
Running six agents means the canvas stops answering the one question that matters — who needs you right now. Here's how Fleet's state grouping and the pending-question tail fix that.

Best-of-N Coding: Ensemble Exploration and Auto-Critique
One task, N divergent forks, a critic that scores them, and a human who picks the winner. When best-of-N earns its token cost, and when it doesn't.

- News
Mission Control: Know Which Agent Needs You
A right-edge panel that groups every agent by state and shows the pending question for anything waiting on you — no more hunting the canvas for a stuck agent.

Why We Built an Orchestrator, Not More Routing Rules
Static trigger/transform rules scale badly once your canvas grows past a couple of agents. Here's why MadoHub now routes through a real tool-calling loop instead — and when you should still use rules.

- News
Ensemble Exploration: Try N Approaches at Once
Ask MadoAgent to fork a problem into several divergent agents, watch them work side by side, and pick the winner — with an optional critic to help you compare.

- News
Flows: Multi-Agent Templates in One Click
Drop a complete, pre-wired set of agents onto your canvas in one click — peer review, pipeline, or watcher — and save your own setups as reusable templates.

- News
MadoHub Cloud Early Access: Buy Once, Own Forever
MadoHub is a $29 one-time purchase with permanent app access and 1,000 non-expiring Starter Credits.

Why We Built MadoHub with Tauri Instead of Electron
The technical reasoning behind choosing Tauri v2 for a desktop development tool — performance, binary size, and Rust's advantages for system integration.

Intelligent Routing: How AI Agents Talk to Each Other
Routing is the hidden infrastructure of multi-agent systems. Here's how MadoHub makes agent-to-agent communication explicit, inspectable, and configurable.

Building Multi-Agent Workflows with MadoHub
A practical guide to composing multiple AI agents into reliable workflows using the MadoHub canvas.

Under the Hood: How JSONL Session Resolution Powers Agent Routing
A deep dive into how MadoHub reads Claude Code's session logs to extract clean, structured responses for intelligent routing.

- News
Agent Routing Now Supports Custom Providers
Connect any OpenAI-compatible API endpoint to MadoHub tiles. Route between different models and providers on the same canvas.

Why Canvas-Based Agent Orchestration Matters
Traditional chat interfaces force linear thinking. Canvas-based orchestration lets you see, connect, and control multiple AI agents simultaneously.

Spatial Thinking for Software Development
Why arranging your work on a canvas isn't just prettier — it's cognitively better. The case for spatial interfaces in developer tools.

- News
Introducing MadoHub Beta
MadoHub is now available in public beta. Run multiple AI agents on an infinite canvas with intelligent routing between them.

5 Multi-Agent Patterns That Actually Work
Battle-tested patterns for orchestrating multiple AI coding agents, from simple pipelines to self-correcting review loops.

From Single Agent to Multi-Agent: A Practical Migration Guide
Most developers start with one AI agent. Here's how to scale to multiple agents working in parallel — and why it changes everything.

Claude Code vs Cursor vs Copilot: Choosing the Right AI Coding Tool in 2026
An honest comparison of Claude Code, Cursor, and GitHub Copilot — pricing, features, strengths, and when to use each one.

Claude Code Tips: 10 Ways to Be More Productive in 2026
Practical tips for getting more out of Claude Code — from keyboard shortcuts to multi-agent workflows. Updated for 2026.

How to Run Multiple AI Agents at the Same Time
A step-by-step guide to running Claude Code, Codex, and other AI coding agents simultaneously for faster development.

How to Automate Code Review with AI Agents
Set up an automated AI code review pipeline where one agent writes code and another reviews it — with practical examples and configuration.

What Is Agentic Development? A Beginner's Guide
Agentic development means using AI agents that autonomously write, test, and review code. Here's what it is, how it works, and how to get started.
