Senior Data Scientist, Data Flywheel

This is a fully remote position, open to applicants in United States.

📋 Description

• Characterize datasets, including languages, conditions, domains, speakers, quality, and areas that are underrepresented.

• Create and implement active-learning loops to prioritize tasks based on anticipated performance impact.

• Develop human-in-the-loop workflows, tools, and model-assisted processes.

• Construct curated datasets and methodologies for representative benchmarking of model quality in real-world speech.

• Conduct experiments to identify which data strategies enhance outcomes.

• Enhance data consistency, cleanliness, organization, clarity, and accessibility for non-technical teams.

• Establish repeatable, documented pipelines for model adaptation specific to domains and customers.

• Collaborate with research, engineering, and data teams.

• Report directly to the VP of Data Operations.

⛳️ Requirements

• Hands-on experience with real data pipelines and model-related challenges in data science, machine learning, or applied research.

• Proficient in Python and data tools; comfortable with building analyses, scoring, and automation independently.

• Experience in data characterization, data selection, active learning, or related challenges.

• Familiarity with speech/audio or NLP models.

• Proven ability to transform ambiguous and complex data situations into measurable improvements for models or products.

• Capability to create reusable systems for others to operate.

• Excellent communication skills, particularly in conveying complex findings to non-specialist audiences.

• Active engagement with AI tools and ability to articulate the tools utilized and systems developed.

• Direct experience with ASR/TTS, audio data, or multilingual/code-switched data is preferred.

• Familiarity with ensemble labeling, pseudo-labeling, or LLM-assisted annotation is preferred.

• Knowledge of data provenance, PII/GDPR-compliant pipelines, or model-improvement compliance is preferred.

• Experience in building custom or fine-tuned models tailored to specific customers or domains is preferred.

• Comfort in collaborating directly with research and engineering teams on shared infrastructure is preferred.

• Required to be located in the United States.

🏝️ Benefits

• Equity

• Bonus

• 10% annual bonus

• Equal employment opportunity

• Applicant accommodations