For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
Rio Kouseki

Rio Kouseki

AI Training Specialist - Generative AI

USA flag
New York, Usa
$35.00/hrIntermediateScale AILabelbox

Key Skills

Software

Scale AIScale AI
LabelboxLabelbox

Top Subject Matter

No subject matter listed

Top Data Types

ImageImage
Computer Code ProgrammingComputer Code Programming
VideoVideo

Top Label Types

Action Recognition
Audio Recording
Classification
Computer Programming Coding
Data Collection
Emotion Recognition
Evaluation Rating
Fine Tuning
Object Detection
Prompt Response Writing SFT
Question Answering
RLHF
Text Generation
Text Summarization
Tracking

Freelancer Overview

I am a detail-oriented AI training and data specialist with hands-on experience in data annotation, reinforcement learning from human feedback (RLHF), prompt engineering, and AI model evaluation. I have contributed to cutting-edge generative AI projects, including large language models and video generation systems, where I evaluated model outputs for accuracy, safety, and relevance, crafted complex prompts to test model limitations, and provided structured feedback to improve model alignment. My background includes generating and annotating both text and video data, performing side-by-side model comparisons, and identifying harmful or biased content to enhance model guardrails. I am skilled in critical thinking, content quality assessment, and collaboration within strict guidelines, and thrive in both independent and team-based remote environments.

IntermediateEnglish

Labeling Experience

Scale AI

Project Outlier - Aether

Scale AIAudioAudio Recording
Listening to Audio recordings and either translating, adding captions, determing if the dataset is valid.

Listening to Audio recordings and either translating, adding captions, determing if the dataset is valid.

2025
Scale AI

Aether Project - Outlier

Scale AIVideoObject DetectionAction Recognition
Various Tasks are provided each differing in nature and allowing assessors to train the model based upon our responses to the dataset provided.

Various Tasks are provided each differing in nature and allowing assessors to train the model based upon our responses to the dataset provided.

2025
Labelbox

Alignerr

LabelboxComputer Code ProgrammingAction RecognitionFine Tuning
Using Git to upload an existing codebase whose primary language is python to create a prompt, then compare the code output of two differing Models. You are to create iterations until the code produced is production-ready.

Using Git to upload an existing codebase whose primary language is python to create a prompt, then compare the code output of two differing Models. You are to create iterations until the code produced is production-ready.

2025
Scale AI

Aether Project - Outlier

Scale AIImageQuestion AnsweringObject Detection
Various Tasks are provided each differing in nature and allowing assessors to train the model based upon our responses to the dataset provided.

Various Tasks are provided each differing in nature and allowing assessors to train the model based upon our responses to the dataset provided.

2025

Education

C

College of Staten Island

Bachelor of Science, Computer Science

Bachelor of Science
2012 - 2016

Work History

A

Amazon

FC Associate

New York
2019 - 2024