Relevant areas include AI evaluation and testing; evaluation science and measurement; AI and machine-learning engineering; research software engineering; systems and data engineering; language technologies; technical assurance and conformity assessment; and technical governance.
Experience may come from research, industry, standards work or specialist practice. We are as interested in people who can show where an evaluation method fails as in those who can build one.