Key Skills

Software

Data Annotation Tech

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Label Types

Data Collection

Evaluation Rating

Freelancer Overview

I am an advanced AI Red-Teaming and LLM Safety Specialist with extensive hands-on experience in data annotation, adversarial evaluation, and quality assurance for large language models. My expertise includes designing and executing adversarial prompt attacks, stress-testing model guardrails, and conducting detailed policy compliance audits to identify vulnerabilities such as jailbreaks, hallucinations, bias, and policy violations. I have worked on high-impact projects involving RLHF workflows, reward model evaluation, and multilingual risk assessments in both English and Hindi. My technical skills span Python, TensorFlow, PyTorch, Hugging Face Transformers, and embedding-based systems, with a proven record of maintaining 95%+ annotation accuracy in high-volume, quality-sensitive environments. I am passionate about improving AI safety and reliability through structured, reproducible evaluation and targeted feedback, with experience across domains like NLP, retrieval-augmented generation, and biomedical AI.

Labeling Experience

Education

Work History