Founding ML Engineer: Audio AI & Model Evaluation

Cantina Labs; Greater London; On-site; GBP 150,138 - 165,151; Full time

Benefits offered by this job

Competitive salary and generous equity

Medical, dental, and vision insurance

42 days of paid time off including 15.

Parental leave & fertility support

401(k) retirement savings plan

Job summary

Cantina Labs is seeking an experienced Machine Learning Engineer (MLE) focused on evaluating audio models for speech generation and recognition. You will build evaluation pipelines for both development and production, and create automated dashboards to report results.

As a founding member of the evaluation team, you will lead efforts and help shape the growth of the team while collaborating with backend, data, and product groups to improve model performance.

Qualifications

• Experience designing evaluation metrics for speech models.
• Experience running user studies on Mechanical Turk or similar platforms.
• Experience training and fine-tuning models for evaluation.
• Strong statistical knowledge to compare results.
• Proficient in software engineering and programming.

Responsibilities

• Design evaluation pipelines for models in development and production.
• Design user studies for subjective model evaluations.
• Convert requirements into measurable metrics and dashboards.
• Develop automated evaluation dashboards to monitor model performance.
• Train new models to capture new evaluation metrics.
• Collaborate with model, data and product teams to align evaluation results with goals.

Skills

Metrics design

MTurk user studies

Model training & fine-tuning

Statistics & analysis

Engineering & programming

ASR/TTS training

Large-scale ML experience

Tools

ASR training tools

Job description

Cantina Labs is seeking an experienced Machine Learning Engineer (MLE) focused on evaluating audio models for speech generation and recognition. You will build evaluation pipelines for both development and production, and create automated dashboards to report results.

As a founding member of the evaluation team, you will lead efforts and help shape the growth of the team while collaborating with backend, data, and product groups to improve model performance.