Xrs Here

I’m a B.Tech Computer Science & Engineering graduate (Heritage Institute of Technology, 2026) with focused experience in AI output evaluation, data annotation, and model quality assessment. I evaluate LLM responses across text generation, reasoning, summarization, coding, and instruction-following tasks with strong written English and high attention to detail.

From 2024–2026, I completed independent AI output evaluation and data annotation work where I evaluated 500+ LLM-generated responses using structured quality rubrics. I performed side-by-side (SxS) response comparisons, ranking outputs for safety, relevance, tone, and helpfulness—directly aligned with RLHF workflows. I also annotated and labeled training data across multiple task types while maintaining consistent guideline adherence across evaluation batches.

I’m especially careful about model reliability: I identified model failure modes like hallucination, reasoning gaps, and instruction drift, then documented findings clearly for review. Alongside evaluation, I’ve built a machine learning prediction pipeline with Scikit-learn and Pandas, and completed text analysis/NLP experiments using NLTK and Regex—plus multiple Python applications emphasizing OOP, modular design, error handling, and readable documentation.