Luke M

AI Contractor

Google Developers • Freelance

Apr 2026 - Present • 1 mo

Evaluating a new personalization feature for Gemini, assessing how effectively the model uses information from past Gemini conversations, Gmail, Google Search, and YouTube activity to deliver more relevant and helpful responses. This role combines creativity with analytical thinking, requiring the design of prompts based on personal experiences and the evaluation of the model’s personalized responses across factors such as grounding, integration, accuracy, and helpfulness.

Business Analyst

Google • Freelance

Jan 2026 - Apr 2026 • 3 mos

Performing side-by-side (SxS) evaluations on GHD, comparing outputs and assessing quality, accuracy, and relevance. Conducting fact-checking to validate information and identify inconsistencies or inaccuracies. Providing detailed, well-structured justifications to support evaluation decisions. Following established evaluation guidelines and quality standards consistently. Using browser-based tools and extensions to support analysis and review tasks. Executing simple command-line instructions when required to support evaluation workflows. Documenting findings clearly and communicating insights to stakeholders when needed.

Annotator

Remotasks • Freelance

Aug 2020 - Jan 2022 • 1 yr 5 mos

Performed high-quality data annotation and linguistic quality assurance to optimize large-scale AI language models and automated workflows. Data Annotation & Labeling: Reviewed and annotated complex data outputs to evaluate overall response quality, accuracy, and contextual relevance. Fact-Checking & Validation: Verified system-generated information against trusted sources to isolate factual inaccuracies, data contradictions, and logical inconsistencies. Technical Workflows: Utilized browser-based tools, dedicated extensions, and simple command-line instructions to efficiently manage and review backend data evaluation tasks. Justifications & Documentation: Followed strict project guidelines to consistently document detailed, well-structured reasoning for every quality decision and communicated core findings to stakeholders.