What will the state of AI be like by this time next year?
Predictions about AI development over the coming year — CLI agents will dominate, AI generators will become standard, and LLM improvements will plateau.
Lovable is no longer a toy
Adventures in vibe coding non-trivial apps.
Smart is not enough: three criteria to select AI tools
Organizations should evaluate AI tools based on compliance, community support, and team management capabilities — not just productivity gains.
Blind vibe coding: lessons from working with an infernal language
Can Claude write code I don't understand?
The future of AI-powered product management
Please stop autogenerating requirement documents
Undercover Agent: AI-Generated Test Coverage for Legacy Codebases
An experiment in using AI code agents to write all the missing unit tests for an existing codebase. Let's take code coverage to 100%.
Why vibe coding won't replace software engineers
Non-techies just don't want to arm wrestle the machine
Pricing strategies in the era of AI: why hourly billing no longer works
Traditional hourly billing is broken for software agencies — AI enables faster development while introducing new costs, necessitating value-based pricing.
Asking AI Agents to test legacy software (so our engineers don't have to).
An experiment in getting LLMs to write the tests we never had time for.
Vibe coding on an airplane
What can we build when our AI doesn't need internet access?