REI Skills
PROJECT GUIDES

Engineering workflows

Browse workflow steps defined by the upstream project and find the skills recommended at each stage.

Ship a SaaS MVPEnd-to-end workflow to scope, build, test, and ship a SaaS MVP quickly.
  1. Plan the scope

    Convert the idea into a clear implementation plan and milestones.

    Define problem, user persona, MVP boundaries, and acceptance criteria before coding.

  2. Build backend and API

    Implement the core data model, API contracts, and auth baseline.

    Prefer small vertical slices; keep API contracts explicit and testable.

  3. Build frontend

    Deliver the primary user flows with production-grade UX patterns.

    Prioritize onboarding, empty states, and one complete happy-path flow.

  4. Test and validate

    Catch regressions and ensure key flows work before release.

    Use go-playwright when the product stack or QA tooling is Go-based.

  5. Ship safely

    Release with basic observability and rollback readiness.

    Define release checklist, minimum telemetry, and rollback triggers.

Build an AI Agent SystemWorkflow to design, implement, and evaluate a production-ready AI agent.
  1. Define use case and reliability targets

    Choose a narrow use case and measurable quality goals.

    Set latency, quality, and failure-rate thresholds before implementation.

  2. Design architecture and retrieval

    Design tools, memory, and retrieval strategy for the agent.

    Keep retrieval quality measurable and version prompt/tool contracts.

  3. Implement orchestration

    Implement the orchestration loop and production safeguards.

    Start with constrained tool permissions and explicit fallback behavior.

  4. Evaluate and iterate

    Run benchmark scenarios and improve weak areas systematically.

    Use test datasets and failure buckets to guide each iteration cycle.

QA and Browser AutomationWorkflow for robust E2E and browser-driven validation across stacks.
  1. Prepare test strategy

    Define critical user journeys, environments, and test data.

    Focus on business-critical flows and keep setup deterministic.

  2. Implement browser tests

    Automate key flows with resilient locators and stable waits.

    Use go-playwright for Go-native automation projects and Playwright for JS/TS stacks.

  3. Triage failures and harden

    Stabilize flaky tests and establish repeatable CI execution.

    Classify failures by root cause: selector drift, timing, environment, data.

Design a DDD Core DomainWorkflow to model complex domains and implement tactical and evented patterns with explicit boundaries.
  1. Assess DDD fit and scope

    Decide if full DDD is justified and define the modeling scope.

    Document why DDD is needed, where to keep it lightweight, and what success looks like.

  2. Create strategic model

    Define subdomains, bounded contexts, and ubiquitous language.

    Classify subdomains and assign ownership before making implementation decisions.

  3. Map context relationships

    Define context integration patterns, ownership, and translation boundaries.

    Prefer explicit contracts and anti-corruption layers where domain models diverge.

  4. Implement tactical model

    Encode invariants with aggregates, value objects, repositories, and domain events.

    Start from invariants and transaction boundaries, not from tables or endpoints.

  5. Adopt evented patterns selectively

    Apply CQRS, event store, projections, and sagas only where required.

    Use evented patterns where consistency and scale tradeoffs are explicit and accepted.

How to write acceptance evidence

Review complete workflow definitions