Skip to content

← Back to standards

SDLC timeline

Where we are today and what comes next as we standardize AWS, Fly.io, observability, and delivery.

Roadmap and goals

  1. Basic observability

    Sentry.io, Grafana Cloud (Loki + Prometheus), and standard /health / /metrics on Fly APIs—enough signal to know when something is wrong.

  2. In progress

    Uniform and consistent standards

    One place to read how we work—this site, sacred branches, checklists, and repo conventions—so staging and production behave the same way everywhere.

  3. In progress

    Balance AWS and Fly surface areas

    APIs and containers on Fly; AWS only where we must (storage, enterprise, shared platform). Clear lines so teams do not rebuild the same ops twice.

  4. Cross-train

    More than one person can deploy, debug tunnels, and read dashboards per tier—no single-owner bottlenecks.

  5. Build consensus on Brytebridge SDLC patterns

    Agree on branch names, GitHub Actions deploys, Cloudflare tunnels, and go-live checklists across product teams.

  6. Secondary Penn Station test

    Run the full SDLC pattern on a second flagship path (Penn Station) to prove it scales beyond the first adopters.

  7. Build secondary SDLC plan with key rotation

    Document how tokens, tunnel secrets, Fly API keys, and MongoDB Atlas IP access rules (Fly egress allowlists) rotate on a schedule—staging, UAT, and production—without emergency fire drills. Checklist: Rotation inventory.

  8. Establish skills and cross-training

    Cursor skills, runbooks, and paired handoffs so new engineers and auditors can follow the same playbook.

Section titled “Related”