🗺️ Product Roadmap
This document outlines the development history and future plan for Zirelia.
📍 Current State (as of 2026-02-20)
| Component | Technology | Status |
|---|---|---|
| Social Platform | Twitter / X | ✅ Production Ready |
| Image Generation | FLUX 1 via Replicate (remote) | ✅ Production Ready |
| LLM Brain | OpenAI ChatGPT (GPT-4 / GPT-4o-mini) | ✅ Production Ready |
| Memory | ChromaDB (local file) | ✅ Active |
| Instagram / Others | Not implemented | 🔴 Planned (Phase 4) |
| Local Image Gen | Not implemented | 🔴 Planned |
| Control Dashboard | Not implemented | 🔴 Planned (Phase 3) |
🟢 Phase 1: Foundation (Completed) ✅
- Persona Engine: Use LLM (OpenAI) to generate thoughts and personality.
- Visual Identity: LoRA model (Replicate) for consistent face generation.
- Smart Scheduler: Timezone-aware posting (Los Angeles) with holiday awareness.
- Twitter Integration: Automated posting and basic "Safe Mode" engagement.
- Documentation: MkDocs site for setup and guides.
🟡 Phase 2: Infrastructure Hardening (Next Step)
Currently, we use local files for memory (chromadb) and simple loops for scheduling.
* PostgreSQL Migration: Move memory from files to the db container using pgvector.
* Celery Integration: Move replier/liker scripts to the redis queue for better reliability.
* Resource Optimization: Remove unused containers if we decide to stay lightweight.
🟠 Phase 3: Control Center (Dashboard)
Building a UI to manage the bot without using the terminal.
* Web Dashboard: A simple interface (Streamlit or React) to:
* View logs live.
* Manually approve/reject drafted posts.
* Trigger "Force Posting".
* Update persona.yaml settings visually.
🔵 Phase 4: Multi-Platform Expansion
Expanding Sienna's presence beyond Twitter.
* Instagram:
* Post images to Feed.
* Post Stories using the mobile-vertical aspect ratio.
* LinkedIn:
* Professional-tone posts (using a different System Prompt).
* Telegram/Discord:
* Chatbot mode for direct interaction with fans.
🟣 Phase 5: Multimedia Evolution
- Voice: Use ElevenLabs to give Sienna a voice for audio messages.
- Video: Use Remotion or Luma Dream Machine to generate short video clips (Reels/TikToks).
- Real-Time Reactivity: Ability to "see" images in replies (Vision API) and comment on them.