Your site looks AI-generated. Here's how to fix it
The default look is underspecified taste, not a hard model limit. Brief with references, name visitor jobs, then audit the live URL and accessibility tree.
#ai
Posts tagged #ai
Audit the accessibility tree, not only the Lighthouse a11y score
Agents and screen readers share the same semantic layer. A green Accessibility category is not proof that money-page actions have roles, names, and landmarks machines can use.
What to ask before you build an AI agent
A pre-build checklist: job vs chatbot, least privilege, approval gates, evals, cost, when you need WebMCP and MCP, and who monitors the URLs the agent needs.
How does the web dev industry right now differ from 3 or 4 years ago?
FID became INP, AI rewrote coding and discovery, frameworks churned, and agency clients started asking for field data instead of one Lighthouse screenshot. Field notes from the middle of 2026, not a nostalgia list.
How to optimize your content for AI: five research views that disagree (usefully)
GEO papers, answer-engine playbooks, crawl hygiene, zero-click studies, and entity graphs do not agree on the first lever. Agencies need a way to choose the right lens, not another vendor checklist.
European tech sovereignty in 2026: a trend map for developers and agencies
Institutional LLMs, EuroHPC, publisher bot controls, and agent standards are converging. A practical map of what changed, with examples agencies can use in client conversations without turning into policy pundits.
