A Connected AI Workflow Stack
The near-term stack is focused: preserve conversations, carry context forward, turn exported sessions into reusable knowledge, then add a real desktop voice layer.
CinderACE
Export chats, reasoning traces, companion data, images, and audio from 14+ AI platforms. It is the easiest way to build a reusable archive for ember-memory, analysis, or future fine-tuning.
ember-memory
Automatic retrieval hooks, shared collections, and adaptive ranking keep AI sessions from starting cold every time you switch tools or come back later.
CinderACE Sessions
A coming layer for turning exported AI conversations into cleaner, reusable session libraries that can feed CLIs, memory systems, and future analysis workflows.
CinderVOX
A desktop voice layer for AI workflows: TTS, STT, hotkeys, provider switching, and voice cloning in one app. Built to give any AI platform a voice and ears on your desktop without turning it into a SaaS subscription.
Embercore
The original companion platform work is being archived, not abandoned. The future version will be rebuilt with the lessons from CinderACE, ember-memory, and CinderVOX behind it.