Kirsten Rielly

Key Skills

Software

Labelbox

Other

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Task Types

Classification

Evaluation Rating

Prompt Response Writing SFT

RLHF

Text Generation

Freelancer Overview

With 22 years of English teaching experience, I'm an expert in text analysis and writing justifications to explain why one LLM response is more successful or helpful to the user than others. An excellent eye for detail is vital when working on multimodal AI training projects, such as labeling frames in a film or TV series for changes in time, location or character.

Labeling Experience

Image to Text

OtherTextQuestion AnsweringEvaluation Rating

The aim of this Outlier project was to test the model's ability to read images correctly in different categories: reasoning, counting, creative writing, scene understanding, infographics, patterns and extracting data. I created prompts that asked a specific question about the image chosen, followed by testing the model on the client's platform. The model created two responses to answer the prompt, which I labeled by scoring how well it followed the system instructions and the overall truthfulness and helpfulness of each one. I also wrote a detailed justification for each response to explain the scores given. (I worked on this project from November 2024 until March 2025 and was promoted to Reviewer after a month.)

The aim of this Outlier project was to test the model's ability to read images correctly in different categories: reasoning, counting, creative writing, scene understanding, infographics, patterns and extracting data. I created prompts that asked a specific question about the image chosen, followed by testing the model on the client's platform. The model created two responses to answer the prompt, which I labeled by scoring how well it followed the system instructions and the overall truthfulness and helpfulness of each one. I also wrote a detailed justification for each response to explain the scores given. (I worked on this project from November 2024 until March 2025 and was promoted to Reviewer after a month.)

2024 - 2024

Scene Boundaries in Film, Documentaries and TV series

LabelboxVideoSegmentation

This project for Alignerr involved watching films, documentaries and TV episodes in a variety of languages and carefully labeling any temporal, character or location changes on a frame by frame basis. (This was earlier this year in 2025, but this isn't an option in the dropdown selection, so I have chosen 2024 to be able to save the information.)

This project for Alignerr involved watching films, documentaries and TV episodes in a variety of languages and carefully labeling any temporal, character or location changes on a frame by frame basis. (This was earlier this year in 2025, but this isn't an option in the dropdown selection, so I have chosen 2024 to be able to save the information.)

2024 - 2024

KS2 National Testing in English (UK)

OtherImageQuestion Answering

I have worked on this annual marking project with Capita three times since 2023, including recently in 2025. It involves labeling thousands of scanned images of national test answers using a platform called MarkManager, following very prescriptive rules about what is and isn't creditworthy. (The project runs from May to June each year.)

I have worked on this annual marking project with Capita three times since 2023, including recently in 2025. It involves labeling thousands of scanned images of national test answers using a platform called MarkManager, following very prescriptive rules about what is and isn't creditworthy. (The project runs from May to June each year.)

2023 - 2024

Education

Work History