Ellarene C.

Key Skills

Software

Scale AI

Other

Top Subject Matter

AI model evaluation and supervised response ranking for LLM/RLHF-style workflows

Multimodal AI training data annotation (image segmentation and detection)

Multimodal AI training data annotation (audio transcription)

Top Data Types

Image

Video

Text

Audio

Document

Top Task Types

Bounding Box

Segmentation

Object Detection

Data Collection

Evaluation/Rating

Classification

RLHF

Audio Recording

Transcription

Action Recognition

Freelancer Overview

AI Evaluation & Annotation Expert (Multiple Platforms: MERCOR, OUTLIER, MICRO1, ONEFORMA, STELLAR AI). Brings 10+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Bachelor of Science, The Open University (2019) and Product Management Bootcamp, BrainStation (2023). AI-training focus includes data types such as Text, Image, and Audio and labeling workflows including Evaluation, Rating, and Segmentation.

Labeling Experience

Outlier AI - Audio work

Scale AIAudioAudio RecordingTranscription

Listening to audio files and transcribing them in English (US). Reading out loud pre-written responses in a conversation to record natural speaking in English (UK).

2025 - Present

Outlier AI - Text work

Scale AITextClassificationRLHF

Categorising AI responses to pre-written prompts, evaluating them based on accuracy, helpfulness, style, and how well it followed the instructions, and also providing a subjective ranking with written opinion.

2025 - Present

Outlier AI - Video work

Scale AIVideoEvaluation/RatingAction Recognition

Evaluating which one of the two video outputs were the closest in recreating the original video.

2025 - Present

Outlier AI - Image work

Scale AIImageBounding BoxSegmentation

Creating bounding boxes for individual people in images. Painting over backgrounds and/or extra people in one colour to identify what needs to be removed from an image. Also annotating images with labels to identify well-known objects including: people, items of clothing, accessories, animals, cars, brands, shops, landmarks, locations etc. and linking those same objects to similar images available online. Describing new environments for the people in the bounding boxes to be recreated into. Writing prompts to then recreate the original image from the newly created individual images.

2025 - Present

Data Analyst (Monitoring, Evaluation and Learning) — Youth Business International

OtherTextData Collection

Managed high-volume partner data validation and ensured completeness, consistency, and quality of submissions for an entrepreneurial programme dataset. Produced documentation and automated parts of the data workflow to improve data reliability for downstream reporting and analysis. Coordinated with partners to maintain clean, reliable data flows for impact measurement and evaluation use cases.• Validated large volumes of partner data for completeness and consistency.• Automated data workflows using Excel and Power Automate to improve reliability.• Produced structured documentation to support consistent data submission and handling.• Designed and distributed surveys and analysed responses to inform programme measurement.

2021 - 2022

Education

Work History