Insights, tutorials, and updates from our team
Week 6 was supposed to be the close — exit interviews, a playbook, a quiet wind-down. The exit interviews said the quiet part out loud and the energy score told us why. So instead of ending it, we made the game a standing practice: a recurring room whose only job is to put the technical people together with no ticket attached.
Three weeks ago we converted every internal checklist we run more than once a month into a Claude Skill. Eleven went in. Six are still active, two got deprecated, three never shipped. Here's what we kept, what we killed, and the three-question filter we wish we'd used on day one.
Two more experiments delivered. The retrospective build caught the architectural decision the written post-mortem missed; the "explain it to me" challenge worked on three of five attempts, and the two that failed were the bigger lesson. Plus the second precondition for replicability we didn't see coming: a team willing to publish its own seams.
Two weeks, two experiments. Design and PM joined for a cross-role night that reshaped how we talk about UX, and a structured build challenge inside Minecraft surfaced architecture issues no sprint review ever did. Plus the harder question we're now sitting with: is any of this replicable?
I completed a full database coursework in just one evening using Claude - not by outsourcing the work, but by accelerating every step of the process. This is a story about how AI shifts your role from writing everything manually to effectively orchestrating the outcome.
What if your laptop had an AI living inside it, silently watching every process, file, and network request — stopping threats before they even become threats? The future of antivirus isn't signatures. It's intelligence.
We gave our developers a shared Minecraft server to unwind after work — and it accidentally became our most productive idea space. Here's what happened in week one of our Game-Test experiment.
RAG (Retrieval-Augmented Generation) lets AI answer questions using your own data — not just general training. Learn how it works and why it's the most practical AI investment you can make right now.
