Key Skills

Software

AWS SageMaker

Appen

Argilla

CrowdSource

Data Annotation Tech

Deep Systems

Google Cloud Vertex AI

HiveMind

Lionbridge

Mercor

Scale AI

Telus

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Geospatial Tiled Imagery

Medical Dicom

Top Task Types

Action Recognition

Audio Recording

Bounding Box

Emotion Recognition

Translation Localization

Freelancer Overview

I have hands-on experience training large language models through multiple platforms including Outlier and Appen, where I contributed to high-quality AI training data across instruction following, reasoning, and safety domains. My work included expert-level data labeling, prompt and response authoring, and rigorous evaluation of model outputs against detailed rubrics. I consistently focused on clarity, factual accuracy, and alignment with policy, while handling edge cases and ambiguous prompts that required nuanced judgment. I’ve also participated in iterative feedback loops, using model error analysis to improve data quality and reduce hallucinations. What sets me apart is my strong understanding of how labeled data directly impacts model behavior. I’m skilled in prompt engineering, comparative ranking, and fine-grained quality assurance, with a track record of meeting tight quality thresholds while maintaining speed and consistency. My experience training models strengthened my ability to think like both a user and a model trainer, anticipating failure modes, improving reasoning depth, and creating data that meaningfully advances LLM performance at scale.

Labeling Experience

Aether

Scale AIVideoBounding BoxClassification

The project focused on improving the performance, safety, and reliability of large language models (LLMs) across a wide range of real-world use cases. The scope included training and evaluating models on tasks related to reasoning, instruction following, content generation, factual accuracy, and policy compliance. I worked with diverse prompts spanning domains such as general knowledge, technical topics, creative writing, customer support, and ethical decision-making.

The project focused on improving the performance, safety, and reliability of large language models (LLMs) across a wide range of real-world use cases. The scope included training and evaluating models on tasks related to reasoning, instruction following, content generation, factual accuracy, and policy compliance. I worked with diverse prompts spanning domains such as general knowledge, technical topics, creative writing, customer support, and ethical decision-making.

2025

Education

Work History