Customer Support Specialist - Freelance AI Trainer

Overview

Mindrift brings together subject-matter experts and project-based AI work for major technology companies. The focus is on evaluating, testing, and helping refine AI systems. This is not a permanent employment role; participation is organized around individual projects.

Please share your CV in English and state your English proficiency level.

What the work may include

• Creating and assessing fee-related questions and statement-clarification cases, including checking whether fees align with published schedules and separating descriptor-mapping issues from actual fraud allegations.
• Developing autopay and payment-setup examples that include common error conditions such as an incorrect payee, not enough funds, a wrong draft date, or skipped confirmation steps.
• Preparing card-replacement and fraud-reporting cases that test urgency around card blocking, recognition of fraud indicators, gathering the right intake details, and handing the case off correctly to disputes.
• Writing account-closure cases that examine retention-offer rules, pending-transaction checks, and whether offers are withheld from customers who are not eligible.
• Reviewing responses for both accuracy and communication quality, including empathy in stressful situations, the right tone for voice or chat, and concise explanations that do not over-promise.

Candidate profile

This project is well suited to people with experience in customer service, banking, or financial services who are comfortable with part-time, non-permanent project work.

Applicants should have a degree in Finance, Business Administration, Economics, Communications, Psychology, Marketing, or a similar field, along with at least 2 years of experience in customer service, retail, banking, or financial services. Recent or current work in customer support, banking, finance, or closely related roles is preferred.

Strong judgment around service boundaries is important, including the ability to read policies or fee schedules and decide which rule applies in a given case. Experience as a QA analyst, call-center trainer, or conversation designer is considered a plus. Strong written English at C1 level or above is required.

How the process works

The workflow is: apply, complete the qualification step(s), join a project, finish tasks, and receive payment.

Time commitment

During active project phases, the expected workload is about 10 to 20 hours per week. This is only an estimate and is not guaranteed, and it applies only while the project is active.

Compensation

Contributors may earn up to the hourly equivalent of $60, depending on skill level and pace of work. Pay can differ from one project to another based on scope, complexity, and the expertise required.

Additional information

Participation is project-based rather than permanent employment.

The role is for a part-time onsite engagement in Arizona, United States.

Submit your application with your English CV and English proficiency level included.

Applicants should be prepared to work on tasks related to customer support scenarios, banking policy interpretation, and AI evaluation.

Estimated effort: 10 to 20 hours weekly during active periods.

Compensation may reach $60 per hour equivalent on this project, with variation across assignments.