Skip to content

IMPLEMENTATION_ROADMAP.md

Latest commit

Β 

History

History
324 lines (229 loc) Β· 11.1 KB

IMPLEMENTATION_ROADMAP.md

File metadata and controls

324 lines (229 loc) Β· 11.1 KB

Gaji Platform: Next Steps

Last Updated: 2025-11-14
Status: Ready for Implementation
Phase: Documentation Complete β†’ Implementation Start

πŸ“Š Current Status

βœ… Documentation Complete (100%)

  • Architecture design finalized (Pattern B - API Gateway)
  • Database strategy defined (PostgreSQL + VectorDB Hybrid)
  • Backend optimization strategies documented (7 strategies)
  • 7 Epics, 37 User Stories fully specified
  • All Korean content translated to English
  • Documentation consolidated (22 files β†’ 12 files)

🎯 Immediate Actions (This Week)

1. Development Environment Setup βš™οΈ

Time: 4 hours
Owners: Both developers

Tasks:

  • Review ARCHITECTURE.md and DEVELOPMENT_SETUP.md
  • Install prerequisites (Java 17+, Python 3.11+, Node 18+, Docker)
  • Clone all repositories (core-backend, ai-backend, frontend, api-contracts)
  • Organize workspace directory structure
  • Start PostgreSQL + Redis via Docker
  • Verify all services start successfully

Success: Health check endpoints responding (Spring Boot :8080, FastAPI :8000)

2. Pattern B Implementation - Backend πŸ”§

Time: 16 hours
Owner: 민영재 (Backend)

Tasks:

  • Create AIProxyController.java with REST proxy endpoints
  • Configure WebClientConfig for async FastAPI communication
  • Implement Circuit Breaker (Resilience4j)
  • Add Redis caching for passage search
  • Write unit tests (>80% coverage)
  • Write integration tests (Spring Boot ↔ FastAPI)

Reference: BACKEND_OPTIMIZATION.md Strategies 1-3

Success: AIProxyController proxies all FastAPI endpoints with tests passing

3. Pattern B Implementation - Frontend πŸ’»

Time: 8 hours
Owner: κ΅¬μ„œμ› (Frontend)

Tasks:

  • Create unified API client (frontend/src/services/api.ts)
  • Remove dual clients (coreApi, aiApi)
  • Update all service layer imports
  • Update SSE EventSource paths to /api/ai/stream/*
  • Update environment variables (VITE_API_URL=http://localhost:8080)
  • Write E2E tests confirming no direct FastAPI calls

Success: Only 1 API client exists, all calls go through Spring Boot :8080

πŸ“… Short-term Goals (Next 2 Weeks)

Week 1: Pattern B Migration Complete βœ…

Deliverables:

  • βœ… AIProxyController implemented
  • βœ… Frontend unified API client
  • βœ… Docker Compose updated (FastAPI internal network only)
  • βœ… All tests passing

Week 2: Epic 0 Infrastructure πŸ—οΈ

Goal: Complete project foundation with pre-processed dataset

Epic File: docs/epics/epic-0-project-setup-infrastructure.md

Stories (39 hours total):

  • 0.1: Spring Boot project setup (API Gateway) - 6h
    • Story: docs/stories/epic-0-story-0.1-spring-boot-backend-setup.md
  • 0.2: FastAPI service setup (internal-only) - 6h
    • Story: docs/stories/epic-0-story-0.2-fastapi-ai-service-setup.md
  • 0.3: PostgreSQL + Flyway migrations (13 metadata tables) - 5h
    • Story: docs/stories/epic-0-story-0.3-postgresql-database-flyway-migrations.md
  • 0.4: Vue.js frontend setup (PandaCSS + PrimeVue) - 6h
    • Story: docs/stories/epic-0-story-0.4-vue-js-frontend-project-setup.md
  • 0.5: Docker configuration (6 services) - 5h
    • Story: docs/stories/epic-0-story-0.5-docker-configuration.md
  • 0.6: Health check endpoints (Pattern B validation) - 8h
    • Story: docs/stories/epic-0-story-0.6-inter-service-health-check.md
  • 0.7: VectorDB data import from pre-processed dataset - 3h
    • Story: docs/stories/epic-0-story-0.7-vectordb-data-import.md

Success: All services running + 10+ novels imported + 100+ characters in VectorDB

πŸš€ Mid-term Goals (Next Month)

Month 1: Core Features (Epic 1-2)

Epic 1: Scenario Foundation (26h - 3 stories)

  • Epic File: docs/epics/epic-1-what-if-scenario-foundation.md
  • Story 1.1: Scenario data model & CRUD API (8h)
    • Story: docs/stories/epic-1-story-1.1-scenario-data-model-api.md
  • Story 1.2: Unified scenario creation modal (12h)
    • Story: docs/stories/epic-1-story-1.2-unified-scenario-creation-modal.md
  • Story 1.3: Scenario validation system (6h)
    • Story: docs/stories/epic-1-story-1.3-scenario-validation-system.md

Epic 2: AI Character Adaptation (32h - 4 stories)

  • Epic File: docs/epics/epic-2-ai-character-adaptation.md
  • Story 2.1: Scenario β†’ prompt engine (10h)
    • Story: docs/stories/epic-2-story-2.1-scenario-to-prompt-engine.md
  • Story 2.2: Context window manager (8h)
    • Story: docs/stories/epic-2-story-2.2-conversation-context-window-manager.md
  • Story 2.3: Multi-timeline character consistency (6h)
    • Story: docs/stories/epic-2-story-2.3-multi-timeline-character-consistency.md
  • Story 2.4: Scenario context testing & refinement (8h)
    • Story: docs/stories/epic-2-story-2.4-scenario-context-testing-refinement.md

Milestone: Users can create scenarios and start AI conversations with Gemini 2.5 Flash

πŸ“ˆ Long-term Goals (Next 3 Months)

Month 2: Discovery & Conversation (Epic 3-4)

Epic 3: Scenario Discovery & Forking (63h - 7 stories)

  • Epic File: docs/epics/epic-3-scenario-discovery-forking.md
  • Story 3.1: Book browse page (8h)
    • Story: docs/stories/epic-3-story-3.1-book-browse-page.md
  • Story 3.2: Book detail page (10h)
    • Story: docs/stories/epic-3-story-3.2-book-detail-page.md
  • Story 3.3: Scenario browse UI & filtering (8h)
    • Story: docs/stories/epic-3-story-3.3-scenario-browse-ui-filtering.md
  • Story 3.4: Scenario forking backend (10h)
    • Story: docs/stories/epic-3-story-3.4-scenario-forking-backend-meta-timeline.md
  • Story 3.5: Scenario forking UI (8h)
    • Story: docs/stories/epic-3-story-3.5-scenario-forking-ui-meta-fork-creation.md
  • Story 3.6: Scenario search & advanced filtering (9h)
    • Story: docs/stories/epic-3-story-3.6-scenario-search-advanced-filtering.md
  • Story 3.7: Social sharing & dynamic og:image (10h)
    • Story: docs/stories/epic-3-story-3.7-social-sharing-dynamic-og-image.md

Epic 4: Conversation System (26h - 3 stories)

  • Story 4.1: Conversation data model & CRUD API (8h)
  • Story 4.2: Message streaming & AI integration (12h)
  • Story 4.3: Conversation forking UI (6h)

Month 3: Polish & Launch (Epic 5-6)

  • D3.js tree visualization
  • User authentication (JWT)
  • Personal memos
  • Beta launch (100 users)

πŸ“š Documentation Structure

Core Documents (Must Read)

Document Purpose
ARCHITECTURE.md Complete architecture guide (6 ADRs)
README.md Project overview
CLAUDE.md AI development guide

Implementation Guides

Document Purpose
DEVELOPMENT_SETUP.md Local environment setup
BACKEND_OPTIMIZATION.md 7 performance strategies
DATABASE_STRATEGY.md Hybrid DB design

Specifications

Document Purpose
PRD.md Product requirements
ERD.md Database schema (13 tables)
API_DOCUMENTATION.md API reference (40+ endpoints)
TESTING_STRATEGY.md Testing guidelines
UI_UX_SPECIFICATIONS.md Design specifications
V0_SCREEN_SPECIFICATIONS.md Screen designs
SECURITY.md Security best practices

Epic/Story Details

  • epics/ - 7 epic specifications
  • stories/ - 37 user story implementations

βœ… Implementation Checklist

Pattern B Migration (40h)

  • Spring Boot AIProxyController (16h)
  • Frontend API client unification (8h)
  • Docker Compose update (4h)
  • Testing & validation (12h)

Epic 0: Infrastructure (39h - 7 stories)

  • 0.1: Spring Boot backend setup (5h)
  • 0.2: FastAPI AI service setup (5h)
  • 0.3: PostgreSQL database + Flyway migrations (5h)
  • 0.4: Vue.js frontend project setup (6h)
  • 0.5: Docker configuration (8h)
  • 0.6: Inter-service health check (5h)
  • 0.7: VectorDB data import (5h)

Epic 1: Scenario Foundation (26h - 3 stories)

  • 1.1: Scenario data model & API (8h)
  • 1.2: Unified scenario creation modal (12h)
  • 1.3: Scenario validation system (6h)

Epic 2: AI Character Adaptation (40h - 4 stories)

  • 2.1: Scenario to prompt engine (12h)
  • 2.2: Conversation context window manager (10h)
  • 2.3: Multi-timeline character consistency (10h)
  • 2.4: Scenario context testing & refinement (8h)

Epic 3: Scenario Discovery & Forking (63h - 7 stories)

  • 3.1: Book browse page (8h)
  • 3.2: Book detail page (10h)
  • 3.3: Scenario browse UI & filtering (8h)
  • 3.4: Scenario forking backend (10h)
  • 3.5: Scenario forking UI (8h)
  • 3.6: Scenario search & advanced filtering (9h)
  • 3.7: Social sharing & dynamic og:image (10h)

Epic 4: Conversation System (26h - 3 stories)

  • 4.1: Conversation data model & CRUD API (8h)
  • 4.2: Message streaming & AI integration (12h)
  • 4.3: Conversation forking UI (6h)

Total Epics 0-4: ~194 hours (~8 weeks for 2 developers)

πŸ“Š Success Metrics

Technical KPIs

Metric Target
API Response Time (P95) < 500ms
AI First Token < 1000ms
Error Rate < 0.1%
Test Coverage > 80%

Product KPIs

Metric Week 2 Month 1 Beta
Novels Ingested 1 10 50
Scenarios Created 5 50 500
Conversations 10 100 1000
Daily Active Users - 10 100

🎯 Action Items (This Week)

민영재 (Backend)

  1. Review ARCHITECTURE.md + BACKEND_OPTIMIZATION.md
  2. Environment setup (Java, PostgreSQL, Redis)
  3. Implement AIProxyController (16h)
  4. Write tests

κ΅¬μ„œμ› (Frontend)

  1. Review ARCHITECTURE.md + UI_UX_SPECIFICATIONS.md
  2. Environment setup (Node.js, pnpm)
  3. Unified API client implementation (8h)
  4. Write E2E tests

Both

  • Daily standup (progress sync)
  • Code review for Pattern B PR
  • Deploy to staging (Railway + Vercel)

Status: Ready to Start Implementation πŸš€
Next Review: End of Week 1 (Pattern B migration complete)
Timeline: 7 weeks to MVP