What does the ops agent look after?
The owner's businesses run self-hosted WordPress sites on Google Kubernetes Engine (GKE), including the taxi booking website that our SEO bot team works on. Self-hosting gives control and flexibility, and it also means someone has to watch pods, databases, disks, certificates and backups. The ops agent does that routine work, and escalates what needs a person.
This runs on our own infrastructure. It is not a client result.What does it monitor?
- Uptime and response: is the site up, and are key pages responding normally?
- Kubernetes health: pods restarting, failing deployments, resource pressure.
- Database health: connections, slow queries, storage.
- Disk and storage: volumes filling up before they cause an outage.
- TLS certificates and domains: upcoming expiries.
- Errors: spikes in server errors or PHP fatal errors.
How are backups and snapshots handled?
The agent takes scheduled backups of WordPress files and databases, and disk snapshots before risky operations, then checks that backups actually completed. Our SEO bots also take backups before content edits. A backup that has never been checked is a hope, not a backup, so restore paths are part of the routine.
What happens during an outage?
For known failure patterns, such as a crashed pod or a full cache, the agent follows predefined recovery steps: restart the workload, clear the specific issue, confirm the site is back, and report what happened. For anything unfamiliar, it gathers diagnostics (logs, recent changes, resource graphs) and alerts the owner in chat with a summary and suggested next steps, instead of improvising.
What does "updating app data" mean?
Our sites show data that changes, such as prices, routes and availability. The agent updates this data through defined, logged processes, for example refreshing route data from a source sheet or database, rather than by editing pages by hand. Combined with our watch routines, it helps make sure what customers see matches what the business intends.
What is the ops agent's stop rule?
- Alone: monitoring, backups and snapshots, verifying backups, documented recovery steps for known incidents, and routine logged data updates.
- With approval: WordPress core, plugin or theme updates, scaling changes that affect cloud costs, database migrations, and restoring from backup over live data.
- Hard stops: deleting data, volumes or clusters; changing DNS or domain settings; changing access, IAM or secrets; anything it doesn't recognise.
- Handover: an alert in chat with diagnostics, what it already tried and recommended next steps.
Why not just use managed WordPress hosting?
For many businesses, managed hosting is the right answer, and we'll say so. Self-hosting on Kubernetes makes sense when you run several sites or apps, need custom components, or want control over cost and data. The trade-off is operational work, which is exactly what this agent reduces. Read what AI agents can't do: complex infrastructure decisions still need a person.
Can you build an ops agent for our sites?
Yes, scoped to your stack, whether that's Kubernetes, a VPS or managed hosting with APIs. We'd start with monitoring, backups and alerting before any automated recovery. See consulting.
What should you prepare before talking to us?
You don't need a technical brief. A useful first conversation needs only a few things, and gathering them is valuable even if you never work with us:
- The process in plain words: what triggers it, who does what, and where it usually goes wrong.
- Rough volumes: how many messages, orders, tickets or invoices per week. A one-week tally is enough.
- Your tools: WhatsApp app or API, CRM, sheets, accounting or helpdesk software, including versions or plans where you know them.
- Three real examples: typical messages or cases, plus one awkward one, with personal details removed.
- What must never happen: the mistakes that would hurt most. These become the first lines of your stop rule.
- An owner: the person who will review the workflow once it runs.
Send these on WhatsApp or by email, or run the pilot scorecard first.
How does a pilot with us work?
Every engagement follows the same four steps. Consult: you walk us through the process, the tools and where it breaks. Pick the pilot: we choose one workflow and write down its inputs, outputs, tools, stop rule and success measure. Ship: we build it into the tools you already use and test it on real cases. Review: your team checks what the agent did and what it escalated, then decides whether to tune it, expand it or stop. Read the full pilot process.