Remote AI Personalization QA Specialist

Turing; Boston (MA); On-site; USD 34,440 - 48,216; Part time

Job summary

A leading AI research accelerator is seeking an AI Quality Analyst to evaluate a new personalization feature for Gemini. The role involves assessing how well AI utilizes your personal data to enhance response relevancy. Applicants should have Russian proficiency, a BS/BA degree in a relevant field, and the ability to work independently and analytically, with flexibility to suit the operational needs. This engagement will last for three months and requires at least 30 hours of work each week with overlaps in PST.

Qualifications

• Ability to read and write in Russian with high proficiency.
• Willingness to use a personal Google account for evaluations.
• Full-time availability in local time zone, overlapping with PST.
• Experience in designing creative, multi-turn prompts for AI.
• Ability to analyze and critique AI responses rigorously.
• Excellent written communication skills for model rankings.
• Self-motivated and able to work independently in a remote setting.

Responsibilities

• Design conversational prompts utilizing personal information.
• Evaluate model responses based on the original intent.
• Analyze responses for grounding and integration quality.
• Stack-rank two model responses for evaluation.
• Write rationales referencing issues in the conversation.
• Maintain strict data hygiene during evaluations.

Skills

Russian Proficiency

Exceptional Analytical Thinking

Creative Prompt Engineering

Strong Evaluation Acumen

Excellent Written Communication

Communication

Independence

Education

BS/BA degree or equivalent experience

Job description

A leading AI research accelerator is seeking an AI Quality Analyst to evaluate a new personalization feature for Gemini. The role involves assessing how well AI utilizes your personal data to enhance response relevancy. Applicants should have Russian proficiency, a BS/BA degree in a relevant field, and the ability to work independently and analytically, with flexibility to suit the operational needs. This engagement will last for three months and requires at least 30 hours of work each week with overlaps in PST.