New·Argos now detects usage deviation across 100+ model endpoints See how →
Home / Blog
Blog

Field notes from production.

Short, practical pieces from the operators in our pods about what is working, what is not, and what it costs.

Sep 2026

Why your AI bill will look like your 2015 cloud bill

Token spend is growing faster than anyone budgeted. The fix is the one FinOps already found: attribution, budgets, anomaly detection and cost per unit of business value.

Aug 2026

The 43-pilot problem: how to stop AI projects without stopping AI

A field guide to triage: what to stop, what to merge, what to accelerate, and how to tell a team their project is over.

Aug 2026

Approve mode is not a waiting room

In regulated work, an agent that recommends and a licensed human who decides is often the destination, not a stage.

Jul 2026

Fix the customer record before you build the agent

Why golden records come first, and what happens to every AI use case built on duplicated master data.

Jul 2026

What a hybrid pod actually looks like on a Tuesday

Who does what when four engineers, a domain expert and six agents share a backlog.

Jun 2026

Staged rollout for agents: shadow, approve, bounded autonomy

Thresholds, kill switches, fairness monitoring, and who signs what.

Let's talk

Tell us the number you need to move.

A 45-minute working session with an operator who has run the kind of work you are describing. You will get an honest read on where your programme stands and what it would take to move it.