What we actually shipped — and what we did not.
Working notes on AI audits, automation, and technical leadership. No hype, no superlatives.
More from the team
Measuring AI ROI: The Metrics That Actually Matter
Most Finnish engineering leaders know what they're spending on AI tools. Almost none can tell you what they're getting back. Here's how to measure AI return on investment in a way that holds up to scrutiny.
When to Build an AI Agent (and When a Simple Automation Is Enough)
Every workflow tool vendor now has an "agent" mode. Every consultant is pitching agents for processes a basic n8n flow would handle in thirty minutes. Here is how to tell the difference before you build the wrong thing.
The Hidden Math Behind Your LLM API Bill
Most teams building LLM features don't notice the cost problem until the invoice arrives. Token budgets, prompt bloat, and naive integration patterns can turn a useful AI feature into an expensive mess — here's what to look at first.
What Nobody Tells You About Running AI Agents in Production
Your AI agent works flawlessly in the demo. Six weeks after deployment it's producing inconsistent outputs, burning through your token budget, and nobody can tell you why. Here's what production reliability for AI automations actually requires.
Before You Commission a Mobile App: What AI-Assisted Development Changes (and What It Doesn't)
Companies getting quotes for mobile apps in 2026 are seeing wildly different timelines and prices. Here's what AI-assisted development actually compresses, where human expertise still decides the outcome, and three questions to ask before you sign anything.
The Faster AI Writes Your Code, the More You Need End-to-End Tests
Vibe coding multiplies the code your team ships while shrinking how deeply humans review it. End-to-end tests are the one safety net that doesn't care who wrote the code — here's why they matter more now, and how to build a suite that runs on every PR.
One letter a month. Nothing more.
What we actually shipped. What we did not. No marketing.