Notes from the studio.
- Sales
Worked example: ai-drafted outbound for a sales team
An illustrative worked example of a research-plus-drafting layer for an assumed 8-rep sales team. Assumed numbers, not a client result; reps send every email.
- Marketing
GPT-5.5 instant is ChatGPT's default: what it means
OpenAI made GPT-5.5 instant the default ChatGPT model with 52.5% fewer hallucinated claims. Here is how a content lead should re-tool the content pipeline.
- Product
AI for product teams: from v0 to a feature that sticks
The path from a Lovable or v0 AI prototype to a production feature users adopt. What to build inside your own surfaces, and what to measure once it ships.
- Marketing
Worked example: a content pipeline for a marketing team
An illustrative worked example of a brief-to-draft content system on Claude and Airtable. Assumed team and numbers, not a client result; an editor signs off.
- Customer support
What Claude Opus 4.8 means for support reply times
Anthropic shipped Claude Opus 4.8 on may 28. Here is what its faster, cheaper drafting means for a support lead: shorter first-response times.
- Finance
AI for invoice processing: cut time-to-pay, no new hire
A finance lead's guide to AI invoice intake. AI reads email and PDF, drafts the coding, and a human approves before anything posts. No new hire, no autopilot.
- Operations
Worked example: claims processing with a human sign-off
An illustrative worked example of AI claims processing for a generic logistics team. Assumed numbers, not a client result; a human approves every claim.
- Governance
Why we ship governance plans with every consultancy engagement
Every AI implementation Stennir ships includes a policy doc, an audit log, and a rollback plan. This is why governance is not an add-on — it is the delivery.