Data Ops Lead

About the role

Turn raw consumer audio streams into clean, reliable training data and build the commercial and operational engine to deliver it to leading AI labs.

Responsibilities

• Own end-to-end journey from raw recordings (500,000+ mobile users) to production-ready datasets for frontier labs
• Structure and manage data deals converting recordings into revenue
• Maintain quality bar for all datasets to retain buyers
• Manage human transcription, annotation, and operations (largely overseas)
• Work directly with CEO on commercial priorities and shape deals
• Interface with buyer-side engineering and research teams to translate specifications into deliverables
• Possess technical fluency in digital audio fundamentals (sample rates, VAD, multichannel formats)
• Partner with internal engineering and external vendors to support pipeline

Requirements

• Authorization to work in the US
• 5+ years building and scaling data pipelines for AI/ML (audio, speech, or multimodal focus)
• Experience delivering data agreements with external partners
• Managing overseas/outsourced teams for data tagging, annotation, QA
• Deep ownership of data quality (QA processes, acceptance criteria)
• Technical fluency in digital audio fundamentals (sample rates, VAD, multichannel formats)
• Experience with training data for TTS, ASR, speaker ID, or full-duplex conversational models
• Familiarity with audio stack (Librosa, FFmpeg, SoX, torchaudio) and cloud data infrastructure (S3, Redshift, BigQuery)
• Understanding of WebRTC-based recording tools for speaker-separated audio capture
• Prior experience leading data or infrastructure teams (hiring, mentoring)