About The Role Alignerr is building a dataset of expert tasks that train and evaluate advanced AI agents on real enterprise work. As a Task Author for the Revenue Operations role, you will design and calibrate realistic problems that challenge AI agents performing the kinds of work a revenue operations professional faces daily — reconciling account data across CRM, orders, email, and chat to arrive at one defensible answer. This is a fully remote, versatile contract role open to experienced RevOps and SalesOps professionals across Sydney and Australia who want to apply their domain expertise to the next frontier of AI development.
Write scoring rubrics that define exactly what a correct answer looks like and how it is evaluated
Set up task environments so an AI agent enters a realistic work situation with the same business systems a human employee would use
Solve each task yourself to validate soundness and confirm the rubric is accurate
Calibrate task difficulty — test tasks against AI models and adjust complexity, ambiguity,
or steps until the task reliably challenges the model to the intended degree while remaining objectively gradable
Review and correct AI-drafted task prompts or rubrics when provided Qualifications Hands-on experience in revenue operations, sales operations, or a closely related function at an industrial or enterprise company
Deep familiarity with CRM systems (e.g., Salesforce) and the ability to query and interpret account and bookings data
Strong understanding of bookings, fulfillment pipelines, and revenue recognition concepts
Ability to define precise, verifiable correctness criteria — not just "good judgment" but a standard that can be checked
Iterative mindset: willing to test, refine, and re-test until a task is correctly calibrated
Comfort operating business communication tools (email, chat) and dashboards as part of multi-system workflows Nice to Have Experience training or evaluating AI tools in a sales or revenue context
Background in enterprise B2B environments with complex deal structures What We Screen For The strongest candidates have done this job in industry, are comfortable with the everyday software of their domain, and can describe — concretely — how they would build a task that reliably challenges a capable AI agent while remaining clearly and objectively gradable.