Hook
Automation reveals what you do not know.
Problem
Manual processes for releases, tests, and infrastructure often depend on hidden steps and tribal knowledge. When you try to automate them, the gaps become obvious.
Why it matters
Automating a workflow forces you to define inputs, outputs, and error handling. This clarity reduces risk and makes work reproducible.
Signals you are here
- Only a few people can complete critical tasks
- Steps are vague or rely on memory
- Processes break when a single person is unavailable
- Automation attempts stall on unclear edge cases
Anti-patterns
- Relying on heroes to execute critical steps
- Skipping documentation because it is hard
- Accepting manual exceptions as normal
- Automation that ignores failure paths
Try this
- Document the workflow with clear inputs and outputs
- Define failure states and recovery steps
- Build small automation probes to validate assumptions
- Replace manual exceptions with deterministic rules
- Add tests that codify the workflow
Example
A release process required a multiplexed SSH session across dozens of terminals. It worked, but only a few senior engineers could run it reliably. The team moved it into a pipeline with explicit steps and ownership, and anyone could release safely.
Reflection prompt
Which manual process feels risky or unclear? Try to automate a small part and see what you learn.
More like this
Heuristic
The Best Configuration Is No Configuration
Defaults beat decisions.
Heuristic
Automate What You Repeat
Repeatable work belongs to code.
Heuristic
Make Infrastructure Disposable
Cattle, not pets.
Heuristic
Increase Contrast, Not Volume
Prompt length does not guarantee novelty. Context contrast does.
Heuristic
State Is Your Enemy, Treat It Carefully
Less state, fewer surprises.
Heuristic
Test Where It Breaks, Not Where It Works
Test the breaks, not the breeze.
