Jarvis Search Relevance & LLM Text Evaluation Project
I worked on a Jarvis Search project aimed at improving the quality of search results on the internet by evaluating AI-generated content. My tasks included annotating and classifying search query results for relevance, conducting LLM evaluation to assess the quality of AI-generated responses, and fine-tuning prompts to optimize AI performance. I also contributed to data labeling for search intent classification, function calling, and text generation tasks to ensure that AI systems deliver highly accurate and contextually relevant answers. This project involved analyzing large sets of structured data (JSON/YAML) and ensuring the alignment of AI behavior with defined standards. Throughout the project, I maintained high quality control standards, identified edge cases, and provided feedback to refine AI models for improved search relevance and user experience.