Protege
DataLab
MachineLearningResearcher-Audio
Neural analysis suggests this role is
optimal for Mid+ candidates.
“Machine Learning Researcher - Audio at Protege. Skills: Audio Data Quality, ML Evaluation, Speech Datasets. Research audio data quality for machine learning. Investigate audio quality effects on model training”
What You'll Achieve.
Establish trustworthy audio-quality baseline; Connect audio-quality issues to dataset improvements; Clearer prioritization over time
Industry & Context.
Root cause analysis
What They're Looking For.
Must Have
Master's degree + 4+ years industry experience, Experience designing and running data evaluations, Experience developing or critically evaluating metrics, Ability to connect low-level signal properties to ML behavior, Comfortable moving between research and production, Excellent written and verbal communication
Nice to Have
PhD or equivalent, Experience with ASR, TTS, speaker modeling, Experience developing evaluation frameworks for training data, Experience inventing, adapting, or validating audio quality metrics, Experience studying dataset quality and model performance, Publications or open-source contributions, Cross-functional collaboration experience, Experience collaborating with industry or academic labs
What You'll Do.
Research audio data quality for machine learning
Investigate audio quality effects on model training
Analyze and summarize Protege's audio catalog
Maintain quality scorecards and metrics
Develop methods to measure acoustic properties
Build workflows for segment-level quality evaluation
Surface localized degradation
Apply quality metrics to detect degradation
Design and run targeted evaluations
Connect audio quality issues to model behavior
Test audio quality metrics correlation
Identify failure modes of metrics
Design better metric alternatives
Translate research into filtering rules
Build scalable tools and pipelines
Track results over time
Make quality signals accessible
and communicate data asset value
How You'll Work.
Team & Collaboration
ML researchers; Data engineers; Data operations; External partners
Communication Scope
Technical docs; Empirical results
Full Job Description
Company Overview: We are building Protege to solve the biggest unmet need in AI — getting access to the right training data. The process today is time intensive, incredibly expensive, and often ends in failure. The Protege platform facilitates the secure, efficient, and privacy-centric exchange of AI training data. Solving AI’s data problem is a generational opportunity. We’re backed by world-class investors and already powering partnerships with some of the most ambitious teams in AI. The company that succeeds will be one of the largest in AI — and in tech. We’re a lean, fast-moving, high-trust team of builders who are obsessed with velocity and impact. Our culture is built for people who thrive on ambiguity, own outcomes, and want to shape the future of data and AI. ROLE OVERVIEW Data is the foundation of AI performance, and we believe model quality starts with data quality. For speech and audio models in particular, the bar for signal fidelity, consistency, and quality control is exceptionally high. We’re seeking a Machine Learning Researcher focused on audio data quality, ML data evaluation, and quality control to lead the evaluation and optimization of large-scale speech datasets used to train audio, speech, and multimodal models. This role will be responsible not only for applying existing audio quality metrics, but also for researching how audio data quality should be evaluated for machine learning systems and developing new methods, benchmarks, and evaluation frameworks that better predict downstream model performance. You will help define what “high-quality audio data” means in the context of modern ML training. That includes studying how different forms of acoustic degradation, dataset inconsistency, recording conditions, speaker variation, labeling quality, segmentation quality, and signal artifacts affect model behavior across ASR, TTS, speaker modeling, representation learning, and multimodal systems. A core part of this role will be original research a
Applying for this Machine Learning Researcher - Audio role?
Most applicants get filtered before a human reads their resume. See if yours makes the cut.
How to Apply on Ashby
- Ashby is a fast modern ATS — most applications take under 3 minutes.
- The resume parser is strong; verify parsed experience dates and job titles.
- Custom screening questions are often scored algorithmically — answer completely.
- Location field affects geo-based screening; use your actual metro area.
ANONYMOUS · UNFILTERED
What do employees actually say about Protege?
Real rants from real employees. Read before you apply.