Guide · Self-reviews

Self-review examples for SRE and DevOps engineers.

If you did the job well, the year looked uneventful from outside. Fewer pages, faster deploys, incidents that ended in minutes. A good SRE self-review turns that calm into a record: the load you took off other people, and the risks you closed before they became incidents.

What reviewers look for in SRE and DevOps

Reliability against a target. SLOs and error budgets, with the numbers.
Less toil. Manual work you automated away, measured in hours or tickets.
Safer change. Deploy frequency, lead time, change failure rate and time to roll back.
Incident follow-through. Blameless reviews, and how many action items actually closed.
Other teams moving faster. Self-serve environments, templates and docs that cut the tickets sent to you.

Example: impact

The projects, people and numbers in these examples are made up. Keep the shape and use your own facts.

Impact • Defined SLOs for the Sync SDK API (99.9% of sync requests succeed over 28 days) with error-budget alerts. Pages for the team dropped from about 30 a week to 8. • Moved deploys from a weekly manual window to on-merge pipelines with a canary stage and automatic rollback. A deploy now takes about 12 minutes instead of 45. • Made staging environment access self-serve with short-lived credentials. Requests that used to wait days now take minutes, and the shared staging password is gone.

The best SRE numbers are about people: pages per week, hours of waiting, tickets that no longer get filed.

Example: collaboration

Collaboration • Facilitated five incident reviews, all blameless, each with owners and dates. 14 of 17 action items are closed; the other three are scheduled. • Paired with the Recipe App 2.0 team on their first dashboards and alerts before launch, so their own service pages them. • Wrote the new-service template with Neha. New services start with logging, metrics and a runbook stub.

Example: growth

Growth • Learned to design reusable Terraform modules instead of copying folders. Three teams use the network module now. • Joined the incident commander rota and asked for feedback after each incident. My status updates got shorter and more regular. • Next half: monthly capacity and spend reviews for the sync cluster.

Phrases that land

"99.9% of sync requests over 28 days": a target anyone can check against the graph.
"pages from about 30 a week to 8": the human cost, reduced.
"requests that waited days now take minutes": toil measured by the people who felt it.
"14 of 17 action items closed": follow-through, with the gap admitted.

Phrases that don't

"Kept the lights on." Everyone hopes so. What did you change?
"Lots of firefighting." It sounds heroic and reads as a system nobody fixed.
"Managed the infrastructure." Which parts, and what got better?
"Was on call." Say what on-call taught you and what you changed after it.

Gather your evidence

With infrastructure as code, your changes are in git too. This lists your commits to infrastructure folders:

git log --since="6 months ago" --author="$(git config user.email)" --no-merges \ --date=short --format='%ad %s' -- terraform k8s .github/workflows

This finds commits that mention incidents, rollbacks or reverts, by anyone. It's a quick index into the hard weeks:

git log --all --since="6 months ago" -i --grep='incident' --grep='rollback' \ --grep='revert' --date=short --format='%ad %an %s'

Then export the pager history (pages per week, before and after your changes), the deploy pipeline's run history, and links to each incident review you ran or joined.

Let Sill keep the record

SRE work is full of waiting: on access, on approvals, on another team's fix. Sill keeps the record while you work. On the Mac it reads your local git history (commit messages and times, never code). On every device it records the tasks you finish, the releases you ship and the follow-ups you send, with receipts like day 6, asked 3×.

At review time, Sill Pro's Review generator turns months of that into a draft per project: what shipped, commit counts and ticket IDs, and every dependency you waited on. Where Apple Intelligence is available it writes a first-person draft on the device, and nothing leaves it. Commits come only from the Mac app. The last 7 days are free; longer periods need Sill Pro, an optional paid upgrade.

Get Sill on the App Store — for Mac, iPhone, iPad and Apple Watch. The daily standup and its set-time reminder are free.

More self-review examples: backend engineers · frontend engineers · iOS engineers · Android engineers · QA engineers · data engineers · engineering managers · product designers · product managers. Or start with how to write a self-review.