AI Training and Evaluation Specialist, Outlier, Oakland

As part of Outlier’s initiative to train generative AI models, I contribute by crafting and answering computer science-related questions to enhance model accuracy and performance. I evaluate and rank AI-generated code across multiple programming languages, including Python, Java, JavaScript/TypeScript, C++, and Verilog, ensuring adherence to instructions, clarity, and contextual relevance. In this role, I assess AI-generated responses for accuracy, conciseness, and cultural appropriateness while refining prompts and evaluation criteria to improve overall model quality. Leveraging my strong attention to detail, I ensure grammatical accuracy, proper punctuation, and compliance with style guidelines. This freelance opportunity provides a flexible, remote work environment, allowing me to optimize AI models while maintaining schedule autonomy