Practical writing
no content theatre
AI systems, delivery decisions, product operations and the mistakes worth writing down. Written by the senior team between engagements, not by a content calendar.
Customer Signal Triage: From Noise to Roadmap
Support tickets, NPS comments, sales notes and analytics all carry signal. The triage system that turns the volume into prioritisable input.
Incident Postmortems That Produce Actual Learning
Most postmortems end as filed documents. The few that produce real change share a structure: blameless framing, contributing factors, and tracked follow-ups.
SLO Design for Products That Do Not Page Engineers
An SLO either protects sleep or burns it. The difference is in how the objective is defined, the window length, and the budget burn-rate alerts that fire.
On-Call Rotations That Do Not Burn Engineers
An on-call rotation that pages every shift loses engineers. Six structural choices that produce rotations engineers will stay on for years.
CI Pipeline Speed: The Three-Minute Rule
A CI pipeline over three minutes pushes the team out of flow. The four levers that consistently bring slow pipelines back inside the budget.
Code Review at Scale Without Becoming a Bottleneck
Code review either distributes context or stalls every PR. Three structural changes that keep review fast, useful and not unfairly loaded on seniors.
Branching Strategy for Teams That Actually Ship
Trunk-based, GitHub flow, GitFlow, release branches: which one fits which team size, with the lead-time numbers that decide it instead of opinion.
Rollback Budget: The Overlooked Deployment Metric
Most teams measure deploy frequency. Few measure rollback time. Without a rollback budget, every release is a one-way bet and recovery is improvisation.
SOC 2 for Engineering Teams: The Evidence Plan
SOC 2 is less about new controls and more about evidence the controls already exist. The engineering work that turns 'we already do this' into a passing audit.
Secret Rotation Without an Outage
Rotating secrets sounds simple until production breaks because two services hold different keys. The dual-key pattern that rotates with zero downtime.
When the Automation Itself Becomes the Incident
Automations fail silently more often than loudly. The observability and recovery patterns that catch a broken workflow before a customer reports it.
Automating the Internal Workflow Without Creating a Second Job
Most internal automation moves the work from doing the task to maintaining the automation. Patterns that keep maintenance below the original task cost.