Freelance Agent Evaluation Engineer

Mindrift; Milano; On-site; EUR 30,000 - 48,000; Part time

Job summary

Mindrift connects specialists with project-based AI opportunities for leading tech companies. You will contribute to building realistic developer environments, crafting AI evaluation tasks, and writing robust tests.

This is a part-time, remote freelance role focused on QA automation and evaluation of AI coding agents. Ideal candidates have 5+ years in software development and a core stack including Python (FastAPI), React, Docker, Postgres, Kafka, and Redis, with strong testing experience and

Qualifications

• 5+ years of professional software development experience.
• Core stack includes Python (FastAPI), JavaScript/TypeScript (React), Docker, Postgres, Kafka, Redis.
• Experience writing tests (functional and integration) to validate solutions.
• English proficiency at B2+ level.

Responsibilities

• Design tasks from intermediate states of simulated environments.
• Write tests that accept multiple correct solutions and reject incorrect ones.
• Evaluate agent solutions and provide QA feedback for fairness and robustness.
• Iterate on tasks and tests based on feedback to ensure solid evaluation.

Skills

5+ years in software development

Python (FastAPI)

JavaScript/TypeScript (React)

Docker

Postgres

Kafka

Redis

Testing (functional, integration)

English proficiency (B2+)

Education

Master’s Degree in Computer Science / related field

Bachelor’s degree accepted with 5 years experience

Tools

Python (FastAPI)

React

Docker

Postgres

Kafka

Redis

Job description

Please submit your CV in English and indicate your level of English proficiency.

Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.

We're building a dataset to evaluate AI coding agents - how well a model handles real-world developer tasks.

You’ll create challenging tasks and evaluation criteria within realistic simulated environments:

• Build realistic developer environments - a virtual company with codebase, infrastructure, and context (tickets, docs, conversations) that forms a believable development history
• Design tasks from intermediate states of these environments - craft the prompt, define what "solved" means, and ensure the task is solvable by an AI agent
• Write tests that verify agent solutions - accept all valid approaches and reject incorrect ones, neither too strict nor too lenient
• Iterate on tasks and tests based on QA feedback - review agent solutions, analyze failures, and refine until the evaluation is fair and robust

What this is NOT:

• Not data labeling
• Not prompt engineering
• Not writing code from scratch - the agent writes most of the code; you guide and evaluate

What we look for:

• 5+ years in software development
• Core stack: Python (FastAPI), JavaScript/TypeScript (React), Docker, Postgres, Kafka, Redis
• Experience writing tests (functional, integration)
• English proficiency - B2+

Why this is hard:

Frontier models are already good at coding. Creating a task that genuinely challenges the best models is non-trivial. You need to deeply understand where models fail and what scenarios reveal the difference between a good and a bad solution. Tasks have many valid solutions - writing tests that accept all correct solutions and reject incorrect ones is harder than it sounds.

Requirements and benefits

Educational qualifications

• A Master’s Degree in Computer Science, Software Engineering, Data Science / Data Analytics, Artificial Intelligence / Machine Learning, Computational Linguistics / Natural Language Processing (NLP), Information Systems or other related fields.
• Bachelor’s degree is accepted if only candidate has 5 years of experience in the field.

Academic and/or Professional Experience

Candidates should have a minimum of 3 years of professional experience in related roles or domain - specifically for QA-automation/testing or cybersecurity roles.

How it works

Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid

Compensation:

Paid per accepted task. Your rate depends on the qualification tier you reach and how efficiently you complete tasks - up to the equivalent of $40/hr. Because payment is per task, a faster pace raises your effective hourly rate.

Why this freelance opportunity might be a great fit for you?

• Take part in a part-time, remote, freelance project that fits around your primary professional or academic commitments.
• Work on advanced AI projects and gain valuable experience that enhances your portfolio. - Influence how future AI models understand and communicate in your field of expertise.