A Planned Night Under London's Sky: AI, Automation, and Four Deep-Sky Objects

Last night was one of those rare London evenings: clear skies, no moon, and a forecast that held. My automated sky alert fired at 5pm, so by 7:30 I had the Seestar S50 out and aligned. But this time I did something new — I planned the session in advance, with AI help. The Setup The telescope is the ZWO Seestar S50 — a compact smart scope with a 50mm aperture, built-in tracking, live stacking, and a phone app that handles everything. It’s not a serious instrument by amateur astronomy standards, but it punches well above its weight for a grab-and-go in a light-polluted city. ...

March 12, 2026

Never Miss a Clear Night: Automated Sky Alerts for My Seestar S50

I got a Seestar S50 — ZWO’s little smart telescope — about a year ago. The hardware is excellent: point it at anything, tap the app, watch the stacking happen live. But there’s an unavoidable problem with amateur astronomy in London: the weather. M31 — the Andromeda Galaxy, our nearest large neighbour at 2.5 million light-years. Seestar S50, 46 minutes integration, July 2025. Dust lanes visible along the disc; M32 satellite galaxy upper right. ...

March 8, 2026

The Self-Optimisation Loop: Teaching an AI to Improve Itself

Most AI assistants start fresh each session. They read their instructions, follow them, then forget everything. Mine doesn’t. It maintains its own documentation, reviews it daily, and proposes improvements. The Core Idea My AI assistant (ClawdBot) wakes up each session by reading workspace files: AGENTS.md – its operating instructions SOUL.md – personality and voice USER.md – who I am, what I prefer TOOLS.md – local configuration (Telegram topic IDs, RSS feeds, routing rules) MEMORY.md – long-term memory memory/YYYY-MM-DD.md – daily logs These files evolve. When I say “remember this,” it updates them. When it learns something, it writes it down. ...

February 15, 2026

Building a Smart RSS News Digest

I noticed that checking news from different sources had continuously been a drain on my time. Bouncing between The Economist, NYT, Guardian, a few YouTube channels — easily 30+ minutes a day just scanning headlines. I decided to optimize it by creating a daily news digest, focusing on the most important news. The Problem Most RSS readers treat all sources equally and sort by timestamp. This drowns important stories in a flood of recent fluff. ...

February 9, 2026