Senior ML Engineer, Safety — Build Scalable Risk Models

Discord Inc.; San Francisco (CA); Vor Ort; USD 220.000 - 248.000; Vollzeit

Benefits dieser Stelle

Equity

Benefits

Zusammenfassung

Discord is seeking a Senior Machine Learning Engineer to join the Safety ML team in San Francisco, focusing on building and deploying models for user safety, with real-time and batch systems for content understanding and account integrity. You will partner with product, engineering, policy, and legal teams to deliver effective ML solutions.

The role emphasizes production-ready ML systems, collaboration, and a growth mindset, with a base US salary range and equity/benefits.

Qualifikationen

• 4+ years of experience in ML engineering or applied ML roles.
• Strong coding skills in Python and fluency in ML frameworks such as PyTorch, JAX, or TensorFlow.
• Experience building performant ML systems at scale from ideation to production.
• Ability to think from first principles with creativity and pragmatic solutions.

Aufgaben

• Analyze data to identify risk patterns and inform model design.
• Build, iterate on, and evaluate models that detect risk and anomalous behavior.
• Collaborate with product, engineering, policy, and Trust & Safety to deploy ML solutions.
• Curate golden label sets at scale and improve data labeling techniques for training.
• Deploy models into production and monitor performance.

Ausbildung

Bachelor’s or Master’s degree in Computer Science, Machine Learning, Statistics, or a related field

Tools

PyTorch

JAX

TensorFlow

Jobbeschreibung

Discord is seeking a Senior Machine Learning Engineer to join the Safety ML team in San Francisco, focusing on building and deploying models for user safety, with real-time and batch systems for content understanding and account integrity. You will partner with product, engineering, policy, and legal teams to deliver effective ML solutions.

The role emphasizes production-ready ML systems, collaboration, and a growth mindset, with a base US salary range and equity/benefits.