Skip to content

Latest commit

 

History

History
170 lines (126 loc) · 5.14 KB

File metadata and controls

170 lines (126 loc) · 5.14 KB

Developer Methodology Playbook

This is the delivery-team companion to the executive pages in docs/.
Principle: AI accelerates, humans decide.

Who this is for

  • Developers implementing app modernization in App1/ to App4/
  • Technical leads running quality gates and architecture decisions
  • Coaches supporting Copilot adoption during execution

Delivery model (same four phases as executive docs)

Phase Purpose Primary outputs Human checkpoint
1. Discover & Document Capture legacy context clearly LotusApp/documentation/* + schema notes Team confirms baseline is accurate
2. Define Requirements Turn discovery into build scope NewApp/docs/requirements.md + user stories Team approves scope/priorities
3. Plan Conversion Choose architecture and sequence NewApp/docs/architecture.md, conversion-plan.md Team approves implementation plan
4. Execute & Deliver Build modern increments safely Code + tests + decision log Team approves quality before merge

Phase 1: Discover & Document

Goal: Remove ambiguity before coding starts.

Inputs

  • Anonymized screenshots (LotusApp/screenshots/)
  • Field/type definitions (LotusApp/data-schema/)
  • Business workflow notes and role context

Copilot usage

  • Use image + text prompts to extract UI elements and inferred rules
  • Ask Copilot to produce structured docs (screens, workflows, validations, edge cases)
  • Ask Copilot to highlight unknowns and assumptions explicitly

Team responsibilities

  • Correct incorrect assumptions
  • Add domain context AI cannot infer
  • Approve a shared baseline package before moving to Phase 2

Definition of done

  • Main forms/workflows are documented
  • Data entities and relationships are captured
  • Known gaps are listed as open questions

Phase 2: Define Requirements

Goal: Convert context into implementation-ready scope.

Copilot usage

  • Generate first drafts of:
    • Functional requirements
    • User stories with acceptance criteria
    • Non-functional requirements (security, accessibility, performance)

Team responsibilities

  • Prioritize backlog for hackathon scope
  • Mark out-of-scope items for follow-up
  • Validate that requirements map to real operations

Definition of done

  • Approved requirements document in NewApp/docs/requirements.md
  • Prioritized scope list and acceptance criteria
  • Explicit in-scope vs. future-scope boundaries

Phase 3: Plan Conversion

Goal: Commit to a practical architecture and sequencing plan.

Copilot usage

  • Propose architecture options with tradeoffs
  • Draft API boundaries, data model, and UI component breakdown
  • Generate an ordered implementation plan

Team responsibilities

  • Select architecture and reject unsafe/unfit options
  • Align plan to timeline and team capacity
  • Validate test-data strategy (dummy/synthetic only)

Definition of done

  • NewApp/docs/architecture.md and NewApp/docs/conversion-plan.md approved
  • Dependency/order of build tasks is explicit
  • Risk and mitigation notes are documented

Phase 4: Execute & Deliver

Goal: Build working increments with safe AI-assisted velocity.

Copilot usage

  • Generate code and tests incrementally (small reviewable chunks)
  • Refactor, explain, and tighten code based on reviewer feedback
  • Help draft docs for design decisions and tradeoffs

Team responsibilities

  • Enforce code review and security checks
  • Verify tests and maintainability
  • Decide merge readiness

Definition of done

  • Working increment in NewApp/src/
  • Test coverage for critical business paths
  • Decisions captured in NewApp/docs/decisions.md

Quality gates by phase

  1. Gate 1 (after Phase 1): "This represents the current system correctly."
  2. Gate 2 (after Phase 2): "This is the scope we agree to build."
  3. Gate 3 (after Phase 3): "This architecture and sequence are approved."
  4. Gate 4 (during Phase 4): "This increment is secure, tested, and maintainable."

No gate pass = no phase progression.


Practical prompt templates

Discovery

Based on these screenshots and schema notes, produce:
1) screen-by-screen workflow,
2) field-level data model mapping,
3) inferred business rules,
4) open questions requiring human confirmation.

Requirements

Generate prioritized user stories with acceptance criteria from this baseline.
Separate must-have, should-have, and future items.

Planning

Propose two architecture options for this scope.
For each: data model, API shape, frontend structure, risks, and why/when to choose it.

Execution

Implement [feature] in small commits:
1) model changes,
2) API endpoint,
3) UI binding,
4) tests.
Explain assumptions and list follow-up tasks.

Suggested hackathon pacing (reference)

Time Focus
9:00–9:15 Setup and scope alignment
9:15–10:00 Phase 1 discovery package
10:00–11:00 Phase 2 requirement refinement
11:00–12:00 Phase 3 architecture and plan
1:00–3:30 Phase 4 build/test/review
3:30–4:30 Demo + handoff + follow-up actions

Use this as a guide; adapt by app complexity.