Introducing Initiativity: When Your AI Assistant Starts Helping First

DMJBot initiativity lets your AI assistant proactively review context, prepare updates, and start helpful work within limits you control.

From Events to Understanding: Perception in AI Harnesses

Updated todo list AI harnesses need perception, not just event processing: the ability to turn scattered signals into a contextual understanding of what is happening.

How Task Delegation Works in DMJBot

DMJBot delegation lets one assistant route work to common subagents, defined specialists, or external CLI agents, keeping the main chat focused while complex tasks run elsewhere.

Local Ollama Models in Your AI Assistant

Audio Input For Your AI Assistant — The Best Way To Catch An Idea

A short post about why voice input matters in the DMJBot mobile app: it lets you quickly capture messy ideas, give the AI more context than typing usually allows, and continue the conversation later from any device.

How North East England and Newcastle Inspire Innovation

A reflection on how Newcastle and the North East of England, shaped by a legacy of practical invention, continue to inspire new ideas in AI, automation, and real-world innovation.

How Much Initiative Should Your AI Assistant Have? A Dial From Wait for Me to Go for It

Should your AI assistant act on its own while you're away? We're exploring a configurable Initiative dial — from Standby to Go-getter — with guardrails like idle-only activation, written policies, and a daily token budget.

Automating my Reddit routine with AI and a web browser

Learn how I automated my daily Reddit workflow with DMJBot using my own account and browser session. See how AI skills, browser automation, and token-efficient page understanding make reliable web automation practical and affordable.

Your AI Dev Team Needs a Manager. DMJBot Is Built for That.

Running multiple AI coding agents across different machines quickly becomes a management problem. DMJBot acts as a central project manager for your AI dev team—coordinating tasks, tracking progress, and letting you control everything from a single chat interface without the chaos of manual orchestration.

How DMJBot Is Different From OpenClaw

DMJBot and OpenClaw aim for similar outcomes, but DMJBot is a lightweight, Docker-based “brain” that connects to devices with user-controlled access rather than taking over a full machine. It also stands out with event-driven automation, where tools can proactively trigger actions instead of waiting for user commands.

A Real Day With DMJBot: It Shipped a Task While I Was Away

This post shows DMJBot handling a real task end-to-end—waiting for a Slack trigger, executing work, and delivering results—without any user involvement. It highlights how event-driven automation turns an AI assistant into a proactive worker rather than a reactive chat tool.

Hello World — Meet DMJBot, Your AI Employee

This post introduces DMJBot from your perspective—an AI assistant that doesn’t just respond, but actually takes work off your plate and keeps things moving while you live your life. It’s about turning scattered tools and constant checking into a simple, reliable system that works for you, even when you’re not there.