Alfonso Correa

Key Skills

Software

Labelbox

OneForma

Remotasks

Other

Scale AI

Top Subject Matter

Psychology expert annotator

LLM Evaluation in spanish

Content writing AI trainer

Top Data Types

Audio

Image

Text

Top Task Types

Prompt Response Writing SFT

Question Answering

RLHF

Translation Localization

Freelancer Overview

I’m a psychologist with an MSc in Behavioral Science in Management from the University of Stirling, with a strong research focus and familiarity with data analysis. Furthermore, I have experience working as a freelance AI trainer within RLHF and LLM projects in language and generalist projects. I have worked with prompt-writing, evaluation and description tasks in both Spanish and English. Moreover, my background as a web developer and content creator for three years equipped me with the knowledge and ability to conduct research and write more than 13 science-based articles on my blog reaching 20000 views per day.

Labeling Experience

Porject Audio description production

LabelboxAudioText Summarization

Describe the voice that you hear in the audio. This voice will have a tone, pitch, accent, gender, age, pace and intentions. Go through the whole audio and mark the variation in the speech quoting the most relevant words.

Describe the voice that you hear in the audio. This voice will have a tone, pitch, accent, gender, age, pace and intentions. Go through the whole audio and mark the variation in the speech quoting the most relevant words.

2024

Quality Evaluation Images

Scale AIImageRLHFEvaluation Rating

Quality eval tasks will include images in the responses to a user prompt, and must be appropriately evaluated on their quality.

Quality eval tasks will include images in the responses to a user prompt, and must be appropriately evaluated on their quality.

2024

Conversation Mode AI Voice Response Rating

Scale AIAudioRLHF

Compare the quality of two different AI-voiced GenAI chatbots. These chatbots are designed to be interacted with through voice rather than text. It will be presented as an audio. Each response that rated and compared in the task is provided as an audio recording of two AI-generated responses being read out loud by a text-to-speech system.

Compare the quality of two different AI-voiced GenAI chatbots. These chatbots are designed to be interacted with through voice rather than text. It will be presented as an audio. Each response that rated and compared in the task is provided as an audio recording of two AI-generated responses being read out loud by a text-to-speech system.

2024

Bulba Reviewers

Scale AITextPrompt Response Writing SFT

The goal as a reviewer is to find or fix as many good tasks as you can, as efficiently as possible. You have three options for what you can do with a review: Accept Fix Reject

The goal as a reviewer is to find or fix as many good tasks as you can, as efficiently as possible. You have three options for what you can do with a review: Accept Fix Reject

2024

Bulba Factuality

Scale AITextPrompt Response Writing SFT

In each task, you will be presented with the response from an AI chatbot to a prompt given by a user. The response will be split into sentences (spans), and the goal is to evaluate the factual accuracy of each of those spans by looking for web resources that support or contradict what is stated by the model.

In each task, you will be presented with the response from an AI chatbot to a prompt given by a user. The response will be split into sentences (spans), and the goal is to evaluate the factual accuracy of each of those spans by looking for web resources that support or contradict what is stated by the model.

2024

Show more

Education

Work History