Niki-rai Tolliver

Key Skills

Software

Appen

Clickworker

Labelbox

Lionbridge

Mercor

OneForma

Telus

Other

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Text

Video

Top Label Types

Action Recognition

Audio Recording

Data Collection

Evaluation Rating

Object Detection

Prompt Response Writing SFT

Transcription

Freelancer Overview

I am an experienced AI trainer and data annotator with a proven track record in multimodal data labeling, preference ranking, and reinforcement learning from human feedback (RLHF). My work includes daily annotation and evaluation of AI outputs—spanning images, audio, and video—using specialized labeling platforms and strict quality guidelines to ensure high accuracy and consistency. I have collaborated with cross-functional AI teams to provide actionable insights for model improvement and have supported large-scale annotation workflows for model fine-tuning and optimization. My background also includes internet safety evaluation, content moderation, and structured feedback for enhancing AI behavior, making me highly adaptable to diverse data domains and fast-paced environments. My attention to detail, technical proficiency, and commitment to delivering reliable training data are at the core of my approach.

Labeling Experience

AI Trainer/ Data Annotator

OtherImageClassificationQuestion Answering

Conducted daily preference ranking of hundreds of multimodal AI outputs (images, audio, and video) against prompts to measure model performance and alignment. • Applied systematic comparative judgment techniques to evaluate relevance, quality, and coherence, ensuring consistent assessment across diverse modalities. • Maintained high accuracy and throughput in large-scale annotation workflows, delivering reliable training data to support model fine-tuning and preference optimization. • Collaborated with cross-functional AI teams to identify quality trends and provide actionable insights for improving generative model capabilities. • Leveraged specialized labeling platforms and strict evaluation guidelines to produce precise annotations, supporting reinforcement learning from human feedback (RLHF) processes.

Conducted daily preference ranking of hundreds of multimodal AI outputs (images, audio, and video) against prompts to measure model performance and alignment. • Applied systematic comparative judgment techniques to evaluate relevance, quality, and coherence, ensuring consistent assessment across diverse modalities. • Maintained high accuracy and throughput in large-scale annotation workflows, delivering reliable training data to support model fine-tuning and preference optimization. • Collaborated with cross-functional AI teams to identify quality trends and provide actionable insights for improving generative model capabilities. • Leveraged specialized labeling platforms and strict evaluation guidelines to produce precise annotations, supporting reinforcement learning from human feedback (RLHF) processes.

2025

AI Data Trainer/ Data Annotator

MercorAudioClassificationQuestion Answering

Conducted daily preference ranking of hundreds of multimodal AI outputs (images, audio, and video) against prompts to measure model performance and alignment. • Applied systematic comparative judgment techniques to evaluate relevance, quality, and coherence, ensuring consistent assessment across diverse modalities. • Maintained high accuracy and throughput in large-scale annotation workflows, delivering reliable training data to support model fine-tuning and preference optimization. • Collaborated with cross-functional AI teams to identify quality trends and provide actionable insights for improving generative model capabilities. • Leveraged specialized labeling platforms and strict evaluation guidelines to produce precise annotations, supporting reinforcement learning from human feedback (RLHF) processes.

Conducted daily preference ranking of hundreds of multimodal AI outputs (images, audio, and video) against prompts to measure model performance and alignment. • Applied systematic comparative judgment techniques to evaluate relevance, quality, and coherence, ensuring consistent assessment across diverse modalities. • Maintained high accuracy and throughput in large-scale annotation workflows, delivering reliable training data to support model fine-tuning and preference optimization. • Collaborated with cross-functional AI teams to identify quality trends and provide actionable insights for improving generative model capabilities. • Leveraged specialized labeling platforms and strict evaluation guidelines to produce precise annotations, supporting reinforcement learning from human feedback (RLHF) processes.

2025

Education

Work History