Notes on AI that actually helps.

What we learn building AI tools for real businesses — written in plain language, with real examples. No hype, no names.

Studio notes

7 September 2026·4 min read

Studio notes — making AI agents reliable in production

This week the agent fleet shipped no big new feature, but three architectural choices spotted across ongoing engagements are worth sharing: caching AI email triage instead of recomputing it on every click, keeping a Qualiopi compliance agent “on a leash” through traceable tool calls, and forcing an insurance coverage check into a strict JSON schema instead of a prose summary. Three transferable lessons for any owner or CIO putting AI agents into production.

Read →

Studio notes

27 July 2026·4 min read

Studio notes — the right data before the agent

This week the agent fleet reworked, for an insurance broker, a policy-comparison agent to keep no file at all, and for a travel agency, turned a decade of scattered spreadsheets into one clean catalog. Two builds, one lesson: know exactly what data your AI touches, and give it the minimum, cleanest version of it.

Read →

Studio notes

20 July 2026·4 min read

Studio notes — reliability before automation

This week the agent fleet reinforced, for a training organization, the traceability of a compliance audit, and for a travel agency, the reliability of a booking pipeline. Two builds, one lesson: structure before you automate.

Read →

Studio notes

13 July 2026·4 min read

Studio notes — visibility before you delegate

This week the agent fleet built, for a courier company, an operations cockpit for the mailbox and a monitoring agent that watches, escalates and proposes — but never acts alone. Two builds, one lesson: get visibility before you delegate any action.

Read →

Studio notes

4 July 2026·4 min read

Studio notes — reliability before automation

This week the agent fleet was at work for a courier company, a training organisation and an insurance broker. Three sectors, one recurring lesson: measure and make things reliable before you automate anything.

Read →