Data Infrastructure Engineer for Petabyte-Scale AI Systems

Cohere; United States; Hybrid; USD 140,000 - 170,000; Full time

Benefits offered by this job

Lunch stipend $75

Full health and dental benefits

RRSP matching, 401K, Pension

100% Parental Leave top-up

6 weeks paid vacation

Annual offsite & travel budget

$500 home office stipend

Job summary

Cohere is a leading security-first enterprise AI company building foundation AI models and end-to-end products for real-world problems. You will contribute to the data infrastructure powering AI training, working with researchers and engineers at the frontier of scale.

This role focuses on petabyte-scale storage, Python-centric development, and storage-focused Kubernetes. You’ll transform unstructured data into performant datasets across S3, GCS, and POSIX while leveraging Beam/Spark/Flink in

Qualifications

• 4+ years of data storage infrastructure experience.
• Proficient in Python.
• Kubernetes expertise on storage (PV, CSI drivers).
• Ability to convert unstructured data into performant datasets across S3, GCS, and POSIX.
• Experience with Apache Beam, Spark, or Flink.
• Nice-to-have: BigQuery, Airflow, or dbt.
• Genuine excitement about AI and willingness to work at the edge of knowledge.

Responsibilities

• Work directly on petabyte-scale storage infrastructure, and the networking and performance challenges that come with it.
• Collaborate daily with researchers and engineers who are some of the best in the world at what they do.

Skills

Storage infrastructure

Python

Kubernetes (storage)

S3/GCS/POSIX

Beam/Spark/Flink

BigQuery/Airflow/dbt

AI enthusiasm

Edge computing

Job description

Cohere is a leading security-first enterprise AI company building foundation AI models and end-to-end products for real-world problems. You will contribute to the data infrastructure powering AI training, working with researchers and engineers at the frontier of scale.

This role focuses on petabyte-scale storage, Python-centric development, and storage-focused Kubernetes. You’ll transform unstructured data into performant datasets across S3, GCS, and POSIX while leveraging Beam/Spark/Flink in