
Founded by Christian Safka & Keyu Chen
Christian — 10 years of experience launching software and AI-native products 0→1, ex-TPM at Microsoft and most recently the Head of ML at Tavus (YC S21).
Keyu — Multimodal and computer vision researcher for over 7 years, working across academia and industry.
The founders met at Tavus, and together have led engineering, research, and product teams from seven countries!
After a long experience building conversation AI video/audio, they're incredibly excited to see what these technologies can unlock for human<>human communication.
The Pinch team has felt first-hand how language barriers can actively limit a team’s productivity. High bandwidth communication is vital for collaboration, and while communication technologies have moved from IRC → VoIP → Video Conferencing, the existing translation tools still operate in low-bandwidth mediums (mostly caption transcriptions). This makes conversations difficult because 1) converting audio to text before translation is slower, 2) you lose all information of how something was said, and 3) you’re no longer having a face-to-face meeting.
The knew there had to be a better way to connect across different languages and cultures, so they started building Pinch. Shortly after, they found out how challenging translation truly was. Balancing latency/accuracy for chunk-based audio translation, capturing inflection and tonality per statement, handling culturally specific phrasing, and making a seamless meeting experience are all unsolved problems Pinch is taking on.
They’re currently in public beta with their own meeting platform, supporting live use-cases in personal connections (e.g. meeting in-laws for the first time), as well as business (foreign clients, international sales, inter-team comms.).
