
Audio & Video Data
Liva AI is a data company building the multimodal datasets that make AI feel truly human. We’re building a factory of consumer-facing software designed to capture naturally occurring voice and video data at scale. We believe multimodal models will become the primary interface for human-computer interaction, yet a massive and rapidly growing data gap stands in the way.
We’ve raised a $3M seed round (backed by YC, Amino Capital, CRV, angels from OpenAI and Meta, and more). We’re working with leading AI labs and voice-agent companies and have sold a wide variety of high-demand datasets.
As a member of our engineering team, you’ll own end-to-end systems for collecting, validating, and quality-assuring frontier multimodal data. This includes building data collection products and annotator workflows, as well as scaling the infrastructure and evaluation pipelines.
WHAT YOU’LL DO:
REQUIRED SKILLS:
BENEFITS:
Liva's mission is to make AI look and sound truly human. The AI voices and faces today feel off, and lack the capability to reflect diverse people across different ethnicities, races, accents, and career professions. We’re fixing that by building the world’s richest library of human voice and video data, fueling the next generation of realistic AI.