Dennis K.

AI Safety and LLM Evaluation Specialist with 8+ years of experience across human-feedback workflows, model safety evaluation, behavioral analysis, quality assurance, trust & safety, and AI training data operations. Experienced in evaluating frontier-model outputs for helpfulness, harmlessness, honesty, factuality, reasoning quality, safety, and instruction adherence. Combines hands-on red teaming, model behavior analysis, reviewer calibration, multilingual evaluation, and high-stakes data review with strong written communication and cross-functional collaboration across distributed teams. Background in Sociology and Criminology, together with ongoing graduate study in Artificial Intelligence, supports nuanced judgment in ambiguous, safety-critical, and sensitive contexts. SELECTED PROFESSIONAL ACHIEVEMENTS • Level 5 Calibrator with approximately 93% average performance across 18 evaluated tasks, reflecting consistent rubric interpretation and quality judgment. • Contributed to quality assurance across more than 2 million reviewed images at Focal Systems and completed 324 production-AI reviews with 270 corrections at Goodnotes. •Sustained approximately 12–15 workflow evaluations per day across selected Goodnotes projects while maintaining rigorous quality review, documentation, and escalation standards. • Evaluation experience spans frontier LLMs, conversational and productivity AI, multilingual systems, medical imaging, computer vision, and autonomous-vehicle data. SKILLS AI Safety Evaluation, Frontier AI & LLM Evaluation, Model Behavior Analysis, Red Teaming & Adversarial Testing, Behavioral Edge Case Analysis, Failure Mode Identification, Model Alignment Evaluation, Safety Risk Identification & Escalation, Prompt & Response Evaluation, Pairwise Response Ranking, Human Preference Evaluation, Helpfulness-Harmlessness-Honesty Assessment, Reasoning & Instruction Adherence Analysis, Hallucination Detection & Fact-Checking, Cross-Model Comparative Evaluation, Reviewer Calibration, Inter-Rater Reliability, AI Quality Auditing, Quality Assurance, Rubric & Taxonomy Development, Human-in-the-Loop AI Evaluation, Model Benchmarking, Multilingual & Cross-Cultural AI Evaluation, High-Volume Evaluation Operations, Trust & Safety, Sensitive Content Review, Cross-Functional Collaboration, Written Communication, Project Coordination, Quality Metrics Tracking, Confidentiality & NDA Compliance