Senior ML Platform Engineer (Remote)

Block; New York (NY); Remote; USD 229,000 - 343,000; Full time

Benefits offered by this job

Remote-first workstyle

Job summary

Block is remote-first and builds ML infrastructure to support fraud detection, recommendations, and underwriting. You will shape our ML platform roadmap, mentor engineers, and drive reliable system design across teams.

You will work on feature stores, model hosting, and training infrastructure while collaborating with applied ML groups to productionize models and improve lifecycle tooling.

Qualifications

• 8+ years in software development or ML on internal platforms.
• Hands-on ML infra experience in feature stores or model hosting.
• Understanding ML lifecycle from data gathering to productionizing.
• Experience designing large-scale distributed systems.
• Oncall and incident response in mission-critical services.
• Ability to mentor engineers and communicate trade-offs.
• Experience with major cloud vendors for high-scale use cases.
• Strong verbal and written communication.

Responsibilities

• Design, build, and operate shared ML infrastructure components.
• Lead the design and delivery of large-scale distributed systems.
• Collaborate with ML teams to shape roadmap and reusable capabilities.
• Build reliable services with automated testing and monitoring.
• Participate in oncall rotations and incident reviews.
• Create self-service tooling and documentation for model deployment.
• Mentor engineers and share best practices.

Skills

Software development

Machine learning infra

Distributed systems

MLOps

Cloud platforms

Mentoring

Oncall experience

Communication

Tools

Kotlin/Java

Golang

Python

TypeScript

React

gRPC/Protobuf

AWS

Databricks

NVIDIA Triton

Job description

Block is remote-first and builds ML infrastructure to support fraud detection, recommendations, and underwriting. You will shape our ML platform roadmap, mentor engineers, and drive reliable system design across teams.

You will work on feature stores, model hosting, and training infrastructure while collaborating with applied ML groups to productionize models and improve lifecycle tooling.