Practical tips and insights about AI in software development.
In spring we compared models and the deck reshuffled every month. Mid-year the picture is calmer — the frontier has largely converged, and price and fit now decide more than raw capability.
Read moreVibe coding is great for prototypes. But for production work, something else wins: write a clear spec, let the agent implement against it, and verify at the right checkpoints.
Read moreOpen-weight models have matured enough that they often just work for everyday dev tasks. No token bill, no code leaving for a vendor. Here's when self-hosting makes sense — and when it doesn't.
Read moreLeadership asks 'is the AI subscription worth it?' and reaches for lines of code or number of prompts. Those are vanity metrics. Here's what to measure instead.
Read moreChatting with AI is nice but limited. The real value shows up when the agent can reach into your systems — Jira, the database, logs, internal docs. That's exactly what MCP is for.
Read moreStop tuning prompt wording. In 2026 the bottleneck isn't how cleverly you phrase it — it's the context you feed the AI. This is the new core skill.
Read morePractical tips on AI for dev teams. No spam, unsubscribe anytime.