AI Coding Challenge Architect

Mindrift; Egypt (PA); On-site; USD 2,755,000 - 4,822,000; Full time

Job summary

Mindrift connects specialists with project-based AI opportunities for leading tech companies. You will design realistic coding tasks, each in a self-contained Docker environment with a broken piece of software that an AI agent must fix, with automated tests verifying the outcome.

You will write a Jira-style instruction document and a reference solve script, calibrate difficulty so most agents fail and some succeed, iterate from QA feedback, and later review other authors' tasks as a QA reviewer.

Qualifications

• 3 years of production software development in one backend stack (Python, Go, Node.js, Java or Rust).
• Fluency in Python pytest and backend testing
• English B2 proficiency

Responsibilities

• Design realistic coding tasks with a broken piece of software in a reproducible Docker environment.
• Build a reproducible Docker environment with pinned dependencies.
• Write a pytest that verifies outcomes, is deterministic, non-flaky, and does not leak the fix.
• Write an instruction markdown that reads like a Jira ticket a developer would receive.
• Write a reference solve script proving the task is solvable.
• Calibrate difficulty so current state-of-the-art agents solve the task 20–60% of the time.
• Iterate based on QA feedback and review other authors' tasks as a QA reviewer.

Skills

Python

Go

Node.js

Java

Rust

pytest

Linux Bash

Tools

Docker

pytest

strace

lsof

Job description

Mindrift connects specialists with project-based AI opportunities for leading tech companies. You will design realistic coding tasks, each in a self-contained Docker environment with a broken piece of software that an AI agent must fix, with automated tests verifying the outcome.

You will write a Jira-style instruction document and a reference solve script, calibrate difficulty so most agents fail and some succeed, iterate from QA feedback, and later review other authors' tasks as a QA reviewer.