AI Benchmark Challenge Architect (Coding/ML/Cyber)

Joinkeywork; United States; Remote; USD 262,000 - 317,000; Full time

Job summary

Joinkeywork is seeking seasoned professionals in coding, ML, or cybersecurity to craft challenging AI benchmark tasks solvable in a Linux terminal. You will convert real problems from your field into clearly defined Linux-based challenges to test cutting-edge agents.

Ideal candidates have extensive engineering or research experience, a master’s degree or higher, at least one publication, and the ability to work from home or remotely under flexible onboarding.

Qualifications

• 8+ years of relevant professional experience in coding, machine learning or cybersecurity.
• Master’s degree or higher; Bachelor with 10 years of experience and a publication may be considered.
• Candidates must have completed their degrees and must not currently be enrolled in a degree program.
• At least one academic or professional publication.
• Practical comfort working in a Linux terminal.

Responsibilities

• Identify difficult, meaningful problems in your area of expertise.
• Translate those problems into tasks that can be solved in a Linux terminal.
• Design challenges that expose limitations in advanced AI agents; test against three models with three attempts per model.

Skills

8+ years experience

Linux terminal

Education

Master’s degree

Bachelor + 10 years + publication

Job description

Joinkeywork is seeking seasoned professionals in coding, ML, or cybersecurity to craft challenging AI benchmark tasks solvable in a Linux terminal. You will convert real problems from your field into clearly defined Linux-based challenges to test cutting-edge agents.

Ideal candidates have extensive engineering or research experience, a master’s degree or higher, at least one publication, and the ability to work from home or remotely under flexible onboarding.