AI Training Specialist (STEM & Coding Expert)
TextRLHF
Performed high-fidelity data labeling and annotation to improve LLM performance in STEM and coding domains. Ensured accurate evaluation and validation of AI-generated code and problem-solving outputs through RLHF and SFT processes. Applied structured annotation frameworks and ethical standards to build reliable datasets for model fine-tuning and alignment.
• Designed, labeled, and validated complex datasets involving mathematical and programming tasks.
• Executed response grading, ranking, and model evaluation within RLHF pipelines.
• Identified and corrected hallucinations to improve safety and adherence to truthfulness criteria.
• Led adversarial red teaming efforts to uncover weaknesses and enhance LLM robustness.
Performed high-fidelity data labeling and annotation to improve LLM performance in STEM and coding domains. Ensured accurate evaluation and validation of AI-generated code and problem-solving outputs through RLHF and SFT processes. Applied structured annotation frameworks and ethical standards to build reliable datasets for model fine-tuning and alignment. • Designed, labeled, and validated complex datasets involving mathematical and programming tasks. • Executed response grading, ranking, and model evaluation within RLHF pipelines. • Identified and corrected hallucinations to improve safety and adherence to truthfulness criteria. • Led adversarial red teaming efforts to uncover weaknesses and enhance LLM robustness.
2025 - Present