Senior Data Scientist, Data Flywheel

Deepgram
United States
Workplace: RemoteFull timeUSD 165,000 - 220,000 annuallyFunction: Data Science & Machine LearningSkills: ["Communication","Experiment design","Automation mindset","Collaboration","Data sensemaking"]

Own the strategy and experiments behind conversational audio data flywheel. Characterize and benchmark complex speech data, design active-learning loops, and create human-in-the-loop workflows that improve model quality faster. Build reusable pipelines and curated datasets that make model performance measurable across real-world diversity, and run rigorous experiments to validate what truly works—turning messy, ambiguous data into repeatable model and product gains.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Deepgram
Deepgram
2 days ago

Senior Data Scientist, Data Flywheel

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 10 hours agoStatus: Live

Job Summary

Own the strategy and experiments behind conversational audio data flywheel. Characterize and benchmark complex speech data, design active-learning loops, and create human-in-the-loop workflows that improve model quality faster. Build reusable pipelines and curated datasets that make model performance measurable across real-world diversity, and run rigorous experiments to validate what truly works—turning messy, ambiguous data into repeatable model and product gains.
Location: United States
Workplace: Remote
Employment Type: Full time
Job Function: Data Science & Machine Learning
Seniority: Mid level

Key Responsibilities

  • •Understand and characterize the data, including languages, conditions, domains, speakers, quality, and underrepresentation.
  • •Design and build active-learning loops to decide what to work on next based on expected performance impact.
  • •Create workflows and model-assisted steps for human-in-the-loop systems, making human attention count.
  • •Build curated datasets and benchmarking methodology to measure model quality across real-world speech diversity.
  • •Run experiments to validate data strategies and turn one-off model work into repeatable, documented pipelines.

Pay and Benefits

Salary: USD 165,000 - 220,000 annually
Equity and Bonus:Equity
Perks:EquityAnnual Bonus

Key Requirements

  • •Hands-on experience working with real data pipelines and model-facing problems in data science, ML, or applied research.
  • •Strong Python and data tooling, with comfort building analysis, scoring, and automation.
  • •Experience with data characterization, data selection, active learning, or related “what to work on next” problems.
  • •Working familiarity with speech/audio or NLP models, including reasoning about output quality, confidence, and error modes.
  • •A track record turning ambiguous, messy data into measurable model or product improvements.
Skills:CommunicationExperiment designAutomation mindsetCollaborationData sensemaking
Tech Stack:PythonNLPASRTTSLLM

Company Brief

Deepgram
Deepgram builds a real-time Voice AI platform delivering speech-to-text, text-to-speech, and voice-agent APIs for developers and enterprises, focusing on low-latency, high-accuracy voice models and scalable deployment options.
Industry: API Platforms
Company Size: Medium (51 to 250 employees)
Growth: Scaleup
Valuation: Unicorn (USD 1B+)
Funding: Series C
Headquarters: San Francisco, United States
Founded: 2015
Glassdoor
Glassdoor: 4.5
WebsiteLinkedInGlassdoor