Loading...

Freelance Agent Evaluation Engineer

  • Part Time
  • Anywhere

Mindrift

We’re building a dataset to evaluate AI coding agents by creating challenging tasks and evaluation criteria within realistic simulated environments. Tasks involve creating virtual companies, assembling and calibrating tasks, designing isolated environments, and writing tests that accept correct solutions and reject incorrect ones. The goal is to create tasks that genuinely challenge the best AI models.

Requirements

  • Degree in Computer Science, Software Engineering, or related fields
  • 5+ years in software development, primarily Python (FastAPI, pytest, async/await, subprocess, file operations)
  • Background in full-stack development, with experience building React-based interfaces (JavaScript/TypeScript) and robust back-end systems
  • Experience writing tests (functional, integration — not just running them)
  • Docker containers, and familiarity with infrastructure tools (Postgres, Kafka, Redis)
  • CI/CD understanding (GitHub Actions as a user: triggers, labels, reading results)
  • English proficiency – B2

Benefits

  • Opportunity to work on AI evaluation projects
  • Flexibility to choose projects and work at your own pace
  • Potential earnings of up to $12 per hour equivalent

Originally posted on Himalayas

To apply for this job please visit himalayas.app.

About this role & career path

Working in Philippines

The Philippines, officially the Republic of the Philippines, is an archipelagic country in Southeast Asia. Located in the western Pacific Ocean, it consists of about 7,641 islands, with a total area of about 300,000 square kilometers, which are broadly categorized in three main geographical divisions from north to south: Luzon, Visayas, and Mindanao. With a population of over 114 million, it is the world's twelfth-most-populous country.

    More jobs at Mindrift

    Keep exploring on Get A Job.ai

    Not quite the right fit? Your next opportunity is a click away.

    Hiring instead? Post a job and reach candidates searching right now.