Deepgram
Senior Data Scientist, Audio
$165K–$220K
USA - Remote · Remote · Senior · Full-time · Posted Sep 21, 2026
About the job
Analyze and characterize conversational audio data, build data-selection and human-in-the-loop systems, and develop representative benchmarks and repeatable pipelines that improve speech model performance.
What you'll do
- Analyze data coverage, quality, and underrepresented areas; design systematic active-learning loops and human-in-the-loop workflows.
- Build curated datasets and benchmarking methods to assess model quality across real-world speech diversity.
- Run experiments and create automated, documented pipelines for model adaptation and data strategy improvements.
What they ask for
- Hands-on experience with real data pipelines and model-facing data science, ML, or applied research problems.
- Strong Python and data-tooling skills, including building analysis, scoring, and automation.
- Experience with data characterization, data selection, active learning, or similar prioritization problems.
- Familiarity with speech/audio or NLP models and ability to assess output quality, confidence, and error modes.
Nice to have
- Direct experience with ASR/TTS, audio data, or multilingual/code-switched data.
- Experience with ensemble labeling, pseudo-labeling, or LLM-assisted annotation.
Pay
$165K – $220K USD base salary
Tools
Python, ASR/TTS, NLP
AI focus
Use AI tools actively to automate and amplify work; build model-assisted human workflows and apply AI-assisted annotation where relevant.
Model-written summary · the original posting is authoritative