Shyan _

I’m a full-stack AI logic specialist combining production-grade backend development with hands-on LLM evaluation. At Turing, I audited AI-generated outputs, validated logic and factual accuracy, and delivered structured justifications using established evaluation rubrics.

I bring a senior reviewer mindset to AI logic validation—catching subtle errors in generated code, spotting security vulnerabilities, and ensuring frameworks like Django and React are used correctly by AI systems. My work emphasizes high inter-rater reliability, clear documentation of findings, and actionable insights that support model improvement cycles.

From TCS, I build and optimize Python/Django backends and JavaScript-driven interfaces for enterprise systems. I designed and maintained RESTful APIs, improved PostgreSQL query throughput by 20–25%, and reduced regression defects by ~30% through pytest-based testing and peer code review.

I also support data and model performance through careful annotation—labeling multimodal datasets with consistent metadata tagging for temporal, action-based, and semantic events. I focus on secure, clean implementation, using input validation, authentication guards, parameterized queries, and output sanitization throughout.