05·The Project

Core Engine Build Status

Epic-level progress tracking for the GamiWays Core Engine build — manually synced from the development repository.

gami-lab/gami-digidouble-core
Overall Progress — Phase A MVP14/27 epics · 52%
14 Done
1 In Progress
12 Planned
PHASE AApril → July 2026

Phase A — Minimal Core

Build and validate the fundamental loop: user input → context assembly → orchestrated avatar response → memory update. Deliver a usable back-office and text-based prototype.

14/21
epics
1.1Platform SetupDone

Monorepo (pnpm + Turborepo), TypeScript strict, Docker Compose, CI baseline

1.2Initial LLM LoopDone

Direct provider SDKs, streaming abstraction, JSON enforcement, retry/timeout wrappers

2.1Conversation LifecycleDone

Session + Conversation entities, message routing, exchange persistence

2.2Persistence LayerDone

PostgreSQL + pgvector schema, Redis session cache, ioredis client

2.3Manual Test ConsoleDone

Back-office Next.js, session inspector, conversation replay, GM debug view

2.4Admin ManagementDone

Scenario config CRUD, avatar management, API key auth

2.5GM Debug ViewDone

Game Master state inspector, directive log, async trigger trace

2.6Runtime InspectorDone

Session-level state visualization, memory snapshot, context assembly trace

O1Health & Dependency MonitoringDone

Health endpoints, dependency checks (DB, Redis, LLM), structured logging

4.1Async Game Master v1Done

Director–Actor model, async trigger evaluation, GM directive injection into avatar flow

4.2cMemory System v3Done

Working memory (sliding window + cumulative summary), episodic persistence, long-term user facts, deterministic selection policy, memory hydration

4.3Performance BaselineDone

Latency measurement (TTFT, end-to-end), token accounting, cost tracking per session/scenario

4.4Multi-Avatar Navigation v1Done

Avatar switching logic, GM-driven transitions, persona isolation per conversation

CX-OBSLLM Observability BoundaryDone

ObservedLlmAdapter enforced across all call sites — Langfuse traces, latency, token counts, cost estimates, model metadata

5.1Knowledge System — Multi-layerIn Progress

Document ingestion (PDF, MD, text), chunking, embeddings, pgvector retrieval, context-aware filtering

5.2Context EnginePlanned

3-dimension context assembly: Memory + Experience/World + Knowledge. Deterministic injection policy.

5.3UX Streaming LayerPlanned

WebSocket streaming, SSE fallback, progressive token delivery to frontend

5.4Guided Progression EnginePlanned

Scenario objectives tracking, progression state machine, GM-driven narrative milestones

6.1Scenario BuilderPlanned

Scenario configuration UI, avatar assignment, knowledge source binding, objective definition

6.2Real Scenario ValidationPlanned

End-to-end test with a real use case (learning or storytelling), user feedback collection

6.3Summer Prototype DeliveryPlanned

Functional prototype: text-based, back-office operational, one real scenario deployed

PHASE BTBD

Phase B — Enhanced Experiences

Add voice input/output, multimedia triggers, multiple scenarios, richer memory systems, and a user-facing frontend.

0/3
epics
B.1Voice Integration (STT + TTS)Planned

STT pipeline (Deepgram / Whisper), TTS streaming (Cartesia / Inworld TTS-2), real-time audio delivery

B.2Multimedia TriggersPlanned

GM-driven media events: image display, video playback, document reveal during conversation

B.3User-Facing FrontendPlanned

Conversation UI, session history, progression visualization, multi-device support

PHASE CTBD

Phase C — Research & Scale Readiness

Prepare the platform for advanced integrations: expressive avatars, advanced persona systems, scaling, SDKs, and research partnerships.

0/3
epics
C.1Expressive Avatar IntegrationPlanned

Real-time video avatar binding, body language triggers, lip-sync coordination

C.2Multi-tenancy & SecurityPlanned

Tenant isolation, JWT auth, RBAC, data residency controls

C.3SDK & API ProductizationPlanned

Public SDK, versioned API, developer documentation, integration examples

Editorial note — This page is manually synced from the development repository gami-lab/gami-digidouble-core . It reflects the actual project state as of the portal's last update date.