Mission
Operations keeps the machine running. The team's job is to handle the analytics, infrastructure, and support functions that aren't anyone else's specialty but everyone depends on.
How they work together
Three lanes: analytics (data instrumentation, reporting, behavioral intelligence), infrastructure (system reliability, scaling, monitoring), and support (process automation and operational responses). The pattern: business-analyst turns metrics into reports, infrastructure-maintainer keeps systems healthy and scaled, operations-optimizer finds and removes process friction, and workflow-analyst maps and automates repeatable work.
Operations sits between every other team — they instrument what Engineering builds, surface signal to Product and Leadership, and reduce manual load across the studio.
Best practices
- Instrument before you launch. business-analyst defines the metrics that matter at design time, not after questions get asked.
- Process is a product. operations-optimizer treats workflows like features — measure baseline, design improvement, ship, measure again.
- Infrastructure is a continuous practice. infrastructure-maintainer monitors proactively. Scaling, drift, and capacity are watched, not waited on.
- Automate the second occurrence. First time, do it manually. Second time, workflow-analyst automates it.
Do's
- Use business-analyst for KPI definition and dashboard design.
- Pull in operations-optimizer any time someone says "this is taking forever every week."
- Let infrastructure-maintainer own scaling decisions — they have the production view.
- Hand new repeatable processes to workflow-analyst for documentation and automation.
Don'ts
- Don't ship dashboards without owners. Every metric needs a human accountable for moving it.
- Don't paper over manual work. Find the structural fix.
- Don't let infrastructure become invisible. Operations gets the credit when nothing breaks.
Common tools
Read, Write, Edit, Grep, Glob, Bash — for scripts, configs, and runbooks. WebSearch / WebFetch for tooling research. Task for orchestrators that delegate process work.
When to call on Operations
Reporting and dashboards, system reliability concerns, scaling challenges, process bottlenecks, or any moment where "this should be more automatic" comes up.