Open gaps
My known gaps, written down with what each one costs before anyone asked. I’d rather show you a gap with a plan than have you find one I hid.
- Open
- 4
- Each row
- Gap · consequence · planned fix
| Gap | Consequence | Fix |
|---|---|---|
| Thin tests on the second siteOn the roadmap | Riflessi’s booking validation, spam gates and webhook retries have no tests, so a change could stop bookings arriving and CI wouldn’t notice. Its one suite covers the fallback email link. | Port DriftPilot’s tests for the same path: the Zod schema, the spam gates, and the webhook client’s retry, backoff and timeout behaviour. |
| No error monitoringOn the roadmap | Neither site alerts anyone when something fails at runtime, with one exception: DriftPilot posts a failed lead in full to Slack. A failed Riflessi booking survives only in a short-lived runtime log, unless the visitor sends the pre-filled email the form offers. | Port DriftPilot’s Slack alert to Riflessi first, then add error monitoring with alerting to both sites. |
| No perf gate on the second siteIn development | Riflessi ships without the Lighthouse budget that guards DriftPilot, so regressions can reach the live site unnoticed. | Port DriftPilot’s config into its CI. I deferred it for launch and wrote down the cost; the port is under way now. |
| No CMS yetOn the roadmap | Content lives in typed accessors in the repository, so a copy change ships as a pull request and a deploy. | Switch on the CMS adapter DriftPilot already keeps behind those accessors. The contract is already there, so only the content source changes. |
Both sites are my own studio’s work. I am looking for a senior role on a team where the standards are shared rather than self-imposed.