Shradha Chandramore

I’m an M.Sc. Genetics postgraduate with 3+ years of combined academic, research, and AI training experience in biology and life sciences. I work as a Biology Subject Matter Expert (SME) evaluating and improving large language model (LLM) outputs using structured scientific review.

In AI training and evaluation, I apply Reinforcement Learning from Human Feedback (RLHF) approaches, perform data annotation and dataset validation, and identify hallucinations, factual inaccuracies, and logical inconsistencies. I design prompt tests to benchmark reasoning and answer quality across models like ChatGPT, Claude, and Gemini.

I also bring a strong experimental biology foundation from research on cytotoxic and genotoxic effects of pesticides using Allium cepa assays, chromosomal aberration analysis, and studies of ziram and aluminium toxicity on plant growth and mitosis. This background helps me validate scientific explanations and convert complex biology concepts into clear, structured training data.

My goal is to strengthen accuracy, reasoning, and reliability in AI systems by combining domain expertise in genetics, cytogenetics, and experimental biology with high attention to detail and quality-focused annotation workflows.