
Behind the Craft Full Tutorial: How to Use Voice AI to Work and Code 10x Faster | Tanay Kothari
108 snips
Nov 23, 2025 Tanay Kothari, CEO and co-founder of Wispr Flow, shares his expertise in voice AI and productivity. He reveals how typing can hinder efficiency, demonstrating how voice can draft emails and code faster. Tanay discusses a groundbreaking brain-to-text prototype that interprets thoughts without speech and emphasizes the importance of habit-driven design over traditional metrics like ARR. He also touches on how Wispr Flow learns users’ writing styles and why he prioritizes retention and user engagement in product development.
AI Snips
Chapters
Transcript
Episode notes
From Silent-Speech Earpiece To Desktop App
- Tanay's team built a silent-speech earpiece prototype that converted thoughts to text after years of research.
- They wrapped that tech into a desktop app and it spread from friends to thousands, prompting a company pivot.
Dictation Models Differ From Transcription
- Wispr Flow built a dictation model that writes what you meant, not literal transcription.
- This product category (dictation vs transcription) requires different models and attention to UI details.
Personalize For Context And Writing Style
- Train the model to learn both short-term and long-term context of each user and app.
- Build personalization so the product 'just gets me' across email, Slack, and other apps.

