Playbooks
How we do it, written down.
Practical guides from our enterprise work on Google Cloud. Free to download: tell us where to send them.
- Playbook 1 · PDFComing soon
Agents in production
How we take an AI agent from prototype to production on Google Cloud, and keep it reliable afterwards.
- Evaluation sets and release gates
- Guardrails and human review
- Deployment, monitoring and SLOs
- Runbooks and on-call
- Playbook 2 · PDFComing soon
Borderless lakehouse implementation
Planning and running a move to open Apache Iceberg tables with one runtime catalog, so every engine works on one copy of your data.
- Assessing your current data estate
- Iceberg table and catalog design
- Governance across engines
- Migration in phases
- Playbook 3 · PDFComing soon
AI cost control
Keeping agent costs predictable at enterprise scale, from prompt design to reporting cost per task.
- Prompt caching
- Model routing
- Cost-per-task reporting in BigQuery
- Budgets and alerts
Get the playbooks
Tell us who you are and we'll send them over. We use your details only to send the playbooks and follow up.