Key Skills

Software

Appen

Google Cloud Vertex AI

OneForma

Label Studio

Top Subject Matter

No subject matter listed

Top Data Types

Document

Image

Text

Top Task Types

Classification

Computer Programming Coding

Data Collection

Object Detection

Text Generation

Freelancer Overview

I am an AI training specialist with hands-on experience in LLM evaluation, prompt testing, and high-quality data labeling across a range of NLP and computer vision tasks. My core strengths include analyzing model responses, assessing reasoning quality, creating optimized prompts, and ensuring accuracy and consistency in text classification, summarization, and dialogue evaluation. I am skilled at following detailed guidelines, applying critical thinking, and delivering precise annotations even for complex or ambiguous cases. In addition to NLP work, I have strong capabilities in image and video annotation, including object detection, scene categorization, and content safety review. I excel at tasks that require strong judgment, attention to detail, and an understanding of AI alignment principles. My goal is to contribute high-quality training data that improves model performance, safety, and reliability across diverse use cases.

Labeling Experience

LLM Response Evaluation & Prompt Quality Analysis for Safety-Aligned NLP Models

AppenTextClassificationQuestion Answering

I worked on a large-scale LLM evaluation project focused on improving model accuracy and safety. My tasks included rating model responses for correctness, reasoning quality, and policy compliance, as well as writing improved prompts and responses for SFT. I also performed red-teaming, content safety classification, and ambiguity resolution. Over the project, I evaluated 10,000+ responses and consistently maintained 95%+ QA accuracy.

I worked on a large-scale LLM evaluation project focused on improving model accuracy and safety. My tasks included rating model responses for correctness, reasoning quality, and policy compliance, as well as writing improved prompts and responses for SFT. I also performed red-teaming, content safety classification, and ambiguity resolution. Over the project, I evaluated 10,000+ responses and consistently maintained 95%+ QA accuracy.

2024 - 2024

Human Activity Video Annotation for Computer Vision Model Training

Label StudioVideoBounding BoxClassification

I worked on annotating short human-activity video clips using Label Studio. My tasks included drawing bounding boxes, tracking objects across frames, and applying action classification labels such as walking, sitting, or picking objects. I annotated over 1,500 video clips and followed strict QA guidelines to ensure consistent and accurate labeling throughout the project.

I worked on annotating short human-activity video clips using Label Studio. My tasks included drawing bounding boxes, tracking objects across frames, and applying action classification labels such as walking, sitting, or picking objects. I annotated over 1,500 video clips and followed strict QA guidelines to ensure consistent and accurate labeling throughout the project.

2024 - 2024

Education

Work History