Faaiza saand

I’m an AI Data Annotator with 2+ years of experience in data annotation, data labeling, and transcription for AI training and LLM development. I apply detailed annotation guidelines and protocols to keep labeled data reliable.

I evaluate and refine AI-generated responses against project quality standards, and I maintain high accuracy through consistent quality assurance and attention to detail. I’m comfortable collaborating with project leads in distributed, remote teams while adapting to evolving annotation criteria.

Across transcription and annotation work, I deliver clean, proofread transcripts with accurate speaker identification and timestamps, while reviewing video clips to identify and categorize recurring events. I manage multiple client projects independently and consistently meet deadlines in remote, deadline-driven environments.

My projects include an event-based video annotation dataset (100+ clips) using Label Studio with a QA pass, an LLM Response Evaluation rubric and dataset (200+ responses) analyzed in Python/Pandas, and a Label Quality vs. Model Performance study using scikit-learn to quantify how annotation errors affect accuracy.