Mandeep K.

Key Skills

Software

Label Studio

Top Subject Matter

No subject matter listed

Top Data Types

Document

Image

Text

Top Task Types

Action Recognition

Data Collection

RLHF

Text Summarization

Translation/Localization

Freelancer Overview

I specialize in training and refining large language models (LLMs) using Reinforcement Learning from Human Feedback (RLHF) methodologies to enhance model performance and alignment with user intent. I can also evaluate AI-generated responses based on user prompts, API calls, and structured rubrics to assess accuracy, relevance, and overall quality. Additonaly, I assign scale ratings (Likert scale) to responses and provide detailed, logical justifications for quality assessments to ensure transparency and continuous improvement. I am also proficient in conducting multi-dimensional evaluations of AI outputs, focusing on factors such as coherence, factual accuracy, and alignment with user needs. I contribute to the iterative refinement of LLMs by providing actionable feedback and insights derived from systematic evaluation processes.

Education

Work History