Devops/SecOps/Software Engineers wanted forAI Benchmarking Task Creators Needed

Description:
We are AirDawg Labs, building an AI benchmark project. We need contributors who can design and debug small technical tasks in Linux environments.
These tasks will be used to test AI agents, so they must be realistic, reproducible, and clearly specified.
What you’ll do:
Create/fix small tasks (broken Dockerfiles, Makefiles, misconfigured environments).
Write task.yaml description (goal, constraints, requirements).
Add at least 6 test cases (simple assertions, run-tests.sh).
Provide a working solution.sh script that passes all tests.
Ensure the task fails initially (null agent must fail) but passes with your solution.
Requirements:
Comfortable with Linux terminal (cd, ls, grep, make, git).
Knowledge of Makefiles and/or Dockerfile debugging.
Ability to write simple test scripts (bash).
Payment:
$20 USD per accepted task.
Tasks must pass review (correct structure, solution passes oracle agent, all tests valid).
Unlimited tasks available — the more you deliver, the more you earn.
Why Join:
Flexible, remote, work-at-your-own-pace.
Great practice for DevOps/debugging skills.
Build a portfolio of reproducible system debugging tasks.
How to Apply:
Please include:
Your experience with Linux/Docker/Makefiles.
Example of a debugging task you’ve solved before (if any).
Estimated number of tasks you can complete per week (The whole project requires 50 accepted tasks within this budget.)