AI Benchmark & Evaluation Engineer - Remote (Part-Time)

Expedock; Philippines; Remote; PHP 482,000 - 827,000; Part time

Job summary

Expedock, a remote-first, tech-enabled workforce platform, seeks a Benchmark & Evaluation Engineer to join on a part-time, hourly basis. You will design and stand up enterprise-like environments across warehouses, lakes, APIs, and DBs in a remote Philippines setting, aligned to a US shift.

You will craft large benchmark datasets with realistic noise and PII, write gold SQL and rubric-scored reasoning keys, and own the evaluation harness to maintain scoring integrity across releases.

Qualifications

• Advanced SQL and data modeling in production environments.
• Strong Python data engineering skills and warehouse loading experience.
• Experience designing evaluation sets and preventing data leakage.
• Ability to build multi-system data environments at scale.

Responsibilities

• Build realistic multi-system data environments across warehouses, lakes, APIs, and DBs.
• Engineer benchmark datasets with realistic noise, seasonality, drift, and synthetic PII.
• Curate benchmark tasks, gold-SQL, and rubric-scored reasoning keys.
• Own the evaluation harness and maintain regression-tracked leaderboards and scoring integrity.

Skills

Advanced SQL

Data Modeling

Python

Pandas

NumPy

Polars

Data warehousing

Benchmark Evaluation

Evaluation design

Tools

Snowflake

BigQuery

Redshift

Postgres

dbt

Airflow

Spark

DuckDB

Job description

Expedock, a remote-first, tech-enabled workforce platform, seeks a Benchmark & Evaluation Engineer to join on a part-time, hourly basis. You will design and stand up enterprise-like environments across warehouses, lakes, APIs, and DBs in a remote Philippines setting, aligned to a US shift.

You will craft large benchmark datasets with realistic noise and PII, write gold SQL and rubric-scored reasoning keys, and own the evaluation harness to maintain scoring integrity across releases.