Remote AI Benchmark & Evaluation Engineer

Motion Ventures; Philippines; Remote; PHP 800,000 - 1,200,000; Full time

Job summary

Expedock is seeking a Benchmark & Evaluation Engineer to design and implement scalable data environments, generate large benchmark datasets, and curate tasks with robust evaluation rubrics. The role requires a production-data mindset and experience with multi-system data warehouses and APIs.

The candidate will own the evaluation harness, maintain regression- tracked leaderboards, and ensure scoring integrity across complex enterprise datasets in a remote setting.

Qualifications

• Proficient in production-grade SQL with advanced features.
• Strong Python for data engineering and data loading pipelines.
• Experience designing evaluation sets with robust validity checks.

Responsibilities

• Build multi-system enterprise environments (warehouses, lakes, APIs, DBs).
• Engineer large benchmark datasets with realistic noise and drift.
• Write and verify benchmark tasks and rubric-scored reasoning keys.
• Own and maintain evaluation harness and leaderboards.

Skills

Advanced SQL

Python programming

Data modeling

Benchmark design

ETL/data pipelines

Education

Bachelor's degree in CS/IS

Tools

Snowflake

BigQuery

Redshift

Postgres

Job description

Expedock is seeking a Benchmark & Evaluation Engineer to design and implement scalable data environments, generate large benchmark datasets, and curate tasks with robust evaluation rubrics. The role requires a production-data mindset and experience with multi-system data warehouses and APIs.

The candidate will own the evaluation harness, maintain regression- tracked leaderboards, and ensure scoring integrity across complex enterprise datasets in a remote setting.