AI Prompt & Policy Specialist

Prompt & Verifier

Remote | Contractor | 3-Month Engagement
20+ Hours/Week | 4-Hour PST Overlap Required

Eligible African Countries: Nigeria, Kenya, Egypt, Ghana

About the Role

We are seeking an AI Prompt & Policy Specialist to help design, evaluate, and improve how AI systems interact with tools, APIs, and users across multiple locales.

This role sits at the intersection of prompt engineering, policy design, AI safety, and operational quality. You will use your understanding of APIs and JSON, SQL fundamentals, policy development, and critical reasoning to review and refine AI prompts, behaviors, and tool workflows.

You’ll work with AI systems interacting with third-party tools and services, ensuring their behavior is accurate, compliant, safe, and user-friendly across different languages and locales.

What You’ll Do

• Design, review, and refine prompts and instructions for AI systems interacting with tools, APIs, and external services.

• Read and interpret API and JSON schemas, arguments, and parameters to verify correct tool usage and behavior.

• Use SQL basics to validate data, investigate issues, perform simple checks, and support analysis of AI or tool behavior.

• Create, document, and maintain policies and guidelines governing AI interactions, content handling, and tool workflows.

• Develop knowledge of specific tool workflows and encode that knowledge into prompts and policies.

• Evaluate typical user actions, workflows, and edge cases across supported tools and services.

• Apply safety and compliance guidelines, including content-filtering requirements.

• Identify harmful, abusive, unsafe, or policy-violating AI outputs.

• Evaluate AI responses for critical reasoning quality, correctness, parameter usage, and policy adherence.

• Validate prompts and AI responses across multiple languages and locales, with attention to nuance, accuracy, and clarity.

• Collaborate with product, engineering, and safety teams to improve prompts, policies, and workflows based on evaluation findings.

• Document edge cases, decision rationales, findings, and recommendations clearly and concisely.

Required Qualifications

• 3–6 years of experience in any combination of:

  • AI operations

  • Content review

  • Quality assurance

  • Product operations

  • Policy

  • Technical support

• Strong API and JSON literacy, including the ability to read schemas, understand arguments, and verify correct parameter usage.

• Working knowledge of SQL, including the ability to run simple queries for data validation and issue investigation.

• Demonstrated experience with prompt engineering or writing clear, structured instructions for AI or automation systems.

• Experience creating, interpreting, and applying policies or operational guidelines.

• Familiarity with tool-specific workflows, including typical user actions and edge cases within SaaS platforms.

• Strong written and verbal communication skills.

• Ability to write clear, concise rationales and explanations.

• Comfort working with multilingual contexts and multiple locales.

• Strong awareness of AI safety and content filtering.

• Ability to identify harmful, abusive, or policy-violating content.

• Excellent critical reasoning and judgment, particularly when evaluating ambiguous or complex cases.

• Exceptional attention to detail, especially when checking parameters, arguments, schemas, and policy requirements.

Core Skills

• SQL

• JSON

• API Literacy

• Prompt Engineering

• AI Policy

• AI Safety

• Content Filtering

• Critical Reasoning

• Quality Assurance

• Policy Evaluation

• Tool Workflows

• Multilingual Evaluation

Engagement Details

• Engagement Type: Contractor assignment

• Duration: 3 months, adjustable based on engagement

• Minimum Commitment: 20 hours per week

• Daily Commitment: At least 4 hours per day

• Required Overlap: 4 hours with PST

• Work Arrangement: Fully remote

• Benefits: No medical or paid leave

Eligible African Countries

• Nigeria

• Kenya

• Egypt

• Ghana

Evaluation Process

• Technical Interview: 60 minutes

What Success Looks Like

Success in this role means consistently producing accurate, safe, policy-compliant, and technically sound AI prompts, evaluations, and workflow recommendations.

You will be expected to identify subtle issues in AI behavior, verify API parameters and JSON structures, apply policies consistently, recognize safety and compliance risks, and clearly document the reasoning behind your decisions across different tools, languages, and user scenarios.