Weave

Quick launch update.

Weave is now live on the App Store — a voice/text journaling app where the AI reflection runs entirely on your phone. No cloud call, nothing synced.

Started as a cloud-AI web app; I've spent the last several weeks rebuilding the whole pipeline on-device instead, because "private journal" should mean the entries never leave the device, not just "we promise not to look."

Stack: Llama 3.2 3B (Q4_K_M) via llama.rn, Whisper small.en via whisper.rn for transcription, MiniLM for semantic embeddings — all local, SQLite, nothing synced anywhere. Metal on iOS, CPU on Android. Works fully in airplane mode after the one-time model download.

Numbers from device testing: iPhone 17 goes save-tap to a complete reflection in ~11.6s. OnePlus 9R on CPU is ~59s — still working on that gap. Reflection quality on a 3B model won't out-nuance GPT-4, and older phones sit right at the memory edge on RAM. Saying that upfront rather than letting someone find it themselves.

iOS only for now (Android when on-device speed catches up). $3.99/mo with a 7-day free trial, no account required:

https://apps.apple.com/us/app/weave-private-ai-journal/id6794576915

If you try it, tell me what's broken — solo dev, and honest feedback shapes what ships next.

If you want to read more about weave: https://www.tryweave.app/about

Quick status update on where this stands.

Weave is a voice/text journaling app where the AI reflection runs entirely on your phone — no cloud call, nothing synced. Started as a cloud-AI web app; I've spent the last several weeks rebuilding the whole pipeline on-device instead, because "private journal" should mean the entries never leave the device, not just "we promise not to look."

Stack: Llama 3.2 3B (Q4_K_M) via llama.rn, Whisper small.en via whisper.rn for transcription, MiniLM for semantic embeddings — all local, SQLite, nothing synced anywhere. Metal on iOS, CPU on Android. Works fully in airplane mode after the one-time model download.

Numbers from device testing: iPhone 17 goes save-tap to a complete reflection in ~11.6s. OnePlus 9R on CPU is ~59s — still working on that gap. Reflection quality on a 3B model won't out-nuance GPT-4, and older phones sit right at the memory edge on RAM. Saying that upfront rather than letting someone find it themselves.

iOS TestFlight is open now: testflight.apple.com/join/wpPKnRXg. Android isn't beta-ready yet — drop in if you want to try the iOS build and tell me what's broken.

For more info: https://www.tryweave.app/essays/why-on-device