Customer Support Specialist - Freelance AI Trainer

Overview

Mindrift brings subject-matter experts into project-based AI work for major technology companies. The focus is on testing, assessing, and refining AI systems. This is not a permanent role; participation is tied to individual projects.

What the work may include

• Developing and reviewing scenarios about fee inquiries and statement clarifications, including checks that amounts align with published fee schedules and that descriptor-mapping questions are separated from genuine fraud reports.
• Creating autopay and payment setup cases with built-in edge cases such as the wrong payee, insufficient-funds policy issues, incorrect draft dates, or skipped confirmation steps.
• Preparing card replacement and fraud-claim scenarios that test urgency in card blocking, recognition of fraud signals, collection of intake details, and proper escalation to the disputes team.
• Writing account-closure cases that assess retention-offer eligibility, pending-transaction review, and whether offers are correctly withheld from customers who do not qualify.
• Assessing replies for both accuracy and communication quality, including empathy, the right tone for voice or chat, and clear wording without making unsupported promises.

Candidate profile

This opportunity suits professionals from customer service, banking, or financial services who are comfortable with short-term, part-time project work. A strong fit is someone who can apply policy judgment and interpret fee schedules or related documents to decide which rule applies in a specific case.

Requirements

• A degree in Finance, Business Administration, Economics, Communications, Psychology, Marketing, or a related area.
• At least 2 years of experience in customer service, banking, retail, or financial services.
• Recent or current work in customer support, banking, financial roles, or similar adjacent positions.
• Good routing judgment and the ability to distinguish between service issues and matters outside service scope.
• Helpful background in QA, call-center training, or conversation design.
• Strong written English at C1 level or above.

How the process works

Applicants go through the following flow: apply, complete one or more qualification steps, join a project, carry out the assigned tasks, and receive payment.

Time commitment

During active project periods, the expected workload is approximately 10 to 20 hours per week. This is only an estimate and is not a guaranteed number of hours.

Compensation

Pay can reach the equivalent of $60 per hour on this project, depending on your experience level and working pace. Compensation differs from project to project based on scope, complexity, and expertise required.

Application note

Please send your CV in English and include your English proficiency level.