Remote ML Platform Engineer: Inference & Training

Coinbase, Inc.; Jackson (MS); Hybrid; USD 191,000 - 225,000; Full time

Benefits offered by this job

Equity

Bonus eligibility

Medical, dental, vision

401(k)

Job summary

Coinbase, Inc. is seeking a Senior Machine Learning Platform Engineer to build and operate the core ML infrastructure that powers feature engineering, model training, and inference across the company.

You will shape streaming, training, and serving systems that enable fraud detection, personalization, and blockchain analysis. You will own high-availability inference, low-latency pipelines, and observability, while mentoring engineers to improve engineering practices and outcomes.

Qualifications

• 5+ years of industry experience as a software engineer with distributed systems in production.
• Built and operated low-latency data or ML infrastructure at scale.
• Experience mentoring engineers and improving engineering quality through reviews.
• Familiarity with ML platform components (feature stores, model serving, training orchestration).
• Uses generative AI responsibly with human oversight to improve workflow efficiency, cost and quality.

Responsibilities

• Own the design and reliability of ML inference infrastructure with high availability.
• Build and optimize low-latency streaming pipelines delivering fresh feature data.
• Enhance distributed training infrastructure for scalable model processing.
• Develop observability tooling to monitor data quality entering models.
• Mentor junior engineers to raise engineering standards.

Skills

Distributed systems ownership

Low-latency infra

Mentoring engineers

ML platform components

Responsible AI

Job description

Coinbase, Inc. is seeking a Senior Machine Learning Platform Engineer to build and operate the core ML infrastructure that powers feature engineering, model training, and inference across the company.

You will shape streaming, training, and serving systems that enable fraud detection, personalization, and blockchain analysis. You will own high-availability inference, low-latency pipelines, and observability, while mentoring engineers to improve engineering practices and outcomes.