Bryan Peña

Key Skills

Software

AWS SageMaker

Clickworker

Remotasks

Toloka

Scale AI

Don't disclose

Any Software

Top Subject Matter

Creative Writing

Geology

Astronomy

Top Data Types

Text

Top Task Types

Data Collection

Prompt Response Writing SFT

Text Generation

Text Summarization

Freelancer Overview

I've answered expert prompts on STEM subjects such as geology and astronomy, crafted prompt + response RAG conversations, and generated creative writing text to train a diverse set of Generative AI applications. I am also experienced in image and video annotation.

Labeling Experience

Image Annotator

Don T DiscloseImageBounding Box

Annotators were tasked to draw bounding boxes around specific types of products or the individuals wearing them. Often, annotators were also required to do their own data collection. In these cases, it was necessary to collect data that was taken from multiple sources, such as from media, in public, in storefronts, in online shops, and others. Quality control was handled by team leaders, approving or disapproving the output one by one. Project size: 12+ months.

Annotators were tasked to draw bounding boxes around specific types of products or the individuals wearing them. Often, annotators were also required to do their own data collection. In these cases, it was necessary to collect data that was taken from multiple sources, such as from media, in public, in storefronts, in online shops, and others. Quality control was handled by team leaders, approving or disapproving the output one by one. Project size: 12+ months.

2023 - 2023

RAG conversation writer

Aws SagemakerText

Created multi-turn RAG conversations to train a generative AI program. Project size: 18 months. Annotators were tasked to make initial edits to their own work, with conversations that did not pass standard sent back to the entire team to edit as a whole.

Created multi-turn RAG conversations to train a generative AI program. Project size: 18 months. Annotators were tasked to make initial edits to their own work, with conversations that did not pass standard sent back to the entire team to edit as a whole.

2023 - 2023

AI Annotator

Don T DiscloseText

Provided expert answers to prompts in a timely manner in order to train a generative AI program. Prompts were primarily in the domain of astronomy, but other STEM topics were involved as well. Overall project size was around 3-5 years, with the possibility of continuing indefinitely, and involved at least three outsourcing houses. Quality measures adhered to were several layers of output review, starting with annotators editing their own work and then submitting it to be reviewed by their teammates and eventually team leaders. Annotators were often required to explain their answers in simple terms to their teammates and team leaders, to prove their familiarity with the subject matter.

Provided expert answers to prompts in a timely manner in order to train a generative AI program. Prompts were primarily in the domain of astronomy, but other STEM topics were involved as well. Overall project size was around 3-5 years, with the possibility of continuing indefinitely, and involved at least three outsourcing houses. Quality measures adhered to were several layers of output review, starting with annotators editing their own work and then submitting it to be reviewed by their teammates and eventually team leaders. Annotators were often required to explain their answers in simple terms to their teammates and team leaders, to prove their familiarity with the subject matter.

2023 - 2023

Education

Work History