Yaps

just setting up my twttr

Pushing this harder, working more and more at the intersect between note taking, offline dictation and knowledge graphs, etc. Keeping everything markdown powered as well - So it easily integrates with other markdown-based tooling. But more importantly, that plus the MCP/CLI, makes it really easy to interact with AI agents, etc. Once I finish optimising some logic stuff behind the scenes, this will be ready to go live.

Shout out to @ElevenLabs for the grant. There was a feature on @TryYaps that I previously hid on Android and the desktop version that may actually be feasible now.

Today I'm launching @TryYaps on Android - a voice-first keyboard I've been rebuilding from the ground up! Ships with everything you'd expect from a modern keyboard - swipe typing, smart suggestions, autocorrect, emoji, GIFs, multi-language, light/dark themes, the whole familiar toolkit. If you're coming from Gboard, nothing to relearn. But the reason it exists is one shortcut away. Dictation that's genuinely 4× faster than typing. It also comes with some robust note taking features. A reading mode that narrates any text aloud. It also cleans up rambling drafts into something send-ready. Every one of those running on-device - no cloud, no listening, no account needed to get started. I kept wondering why no one had just built voice INTO the keyboard itself. The answer is simple, it's tough to do well 😅. Either way I built it, and will keep improving it. Feels a bit unreal to be shipping a full-stack keyboard as one person but it was fun and a necessary first step for a much broader suite of tools. If you've ever want to just yap the thing instead of typing it out - link in the comments.

@tbpn Which solo dev'd app will make it's way onto my phone

We're Back! $1B GLP-1 Lessons. AI Funding Debates. China's 2030 Master Plan. x.com/i/broadcasts/1…

10 AI tools worth knowing right now: (MUST BOOKMARK) 1. Claude → solve complex problems 2. Perplexity → research anything fast 3. Runway → create AI videos 4. Luma AI → generate 3D visuals 5. Suno → make full songs in seconds 6. Gemini → multimodal tasks (images, docs, data) 7. CapCut → edit videos fast 8. NotebookLM → turn content into notes, podcasts, insights 9. Canva → design anything easily 10. Midjourney → generate insane images What am I missing?👇

Big step forward! Shipping @tryYaps to testers today on Android: - Low latency, offline dictation, with a lot less permission requirements than competitors - Seamless text-to-speech (optional) - Note taking, checklists, and Kanban boards integrated Hit me up for access

Just wrapping up some assets for the mobile app store

@claudeai Save time reading

New in Claude Code: auto mode. Instead of approving every file write and bash command, or skipping permissions entirely, auto mode lets Claude make permission decisions on your behalf. Safeguards check each action before it runs.

@jack Welp...