Skip to main content

AI Coding Agent Evaluation Software Developer at toloka-ai

Compensation

$50/hr

Setup
Remote - Australia
Type
Contract
Level
senior
Posted

Description

Mindrift is seeking experienced software developers for project-based work creating and evaluating datasets for AI coding agents. The role involves building realistic simulated developer environments, designing intermediate-state coding tasks and success criteria, writing functional and integration tests, and iterating based on QA feedback to make evaluations fair and robust. The opportunity is not permanent employment or data labeling, requires 5+ years of software development experience, English proficiency at B2+ or higher, and familiarity with Python/FastAPI, JavaScript/TypeScript/React, Docker, Postgres, Kafka, and Redis. Work is self-scheduled and compensated up to $50 per hour equivalent, with tasks estimated at approximately 20 hours each.

For job seekers

Ready to find a role that actually fits?

Upload your résumé, start a Job Search Thread, and let Metaintro rank real openings against your experience — then guide you from search to offer.

Match

Compare live roles against your current evidence.

Position

Turn proof projects into role-specific applications.

Improve

Use market feedback to keep the skill plan current.

Return to navigation