David Osore

Key Skills

Software

Scale AI

CrowdSource

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Label Types

Evaluation Rating

Classification

Freelancer Overview

I am an experienced AI Trainer and Data Analyst with over five years in the tech industry, specializing in data labeling, annotation, and AI training data for natural language processing and conversational AI systems. My background includes selecting, cleaning, and validating diverse datasets to ensure accuracy and unbiased results, as well as designing and refining NLP/NLU models for chatbots and virtual assistants. I am skilled in tools such as SQL Server, Power BI, SAP, and a range of AI technologies including ChatGPT, Jasper, and GitHub Copilot. My work spans from synthesizing and annotating economic and mathematical data to configuring conversational flows and evaluating AI performance, always with a strong focus on attention to detail, collaboration, and ethical judgment. I am passionate about leveraging data to improve AI systems and eager to continue deepening my expertise in this rapidly evolving field.

Labeling Experience

AI Trainer at Henson Group AI Experts

Scale AITextEvaluation Rating

As an AI Trainer, I selected and preprocessed data, ensuring it was clean, unbiased, and relevant for system input. I configured conversational conditions and responses, structuring and refining NLP/NLU models for improved contextual understanding. My work included ongoing evaluation of AI performance and redesign of conversational flows to enhance virtual assistant and chatbot accuracy. • Conducted data selection, cleaning, and preprocessing for conversational AI datasets. • Configured and evaluated AI conversational systems using linguistic and business metrics. • Structured and optimized NLP/NLU models to interpret textual data accurately. • Designed conversational flows for virtual assistant/chatbot usability and error reduction.

As an AI Trainer, I selected and preprocessed data, ensuring it was clean, unbiased, and relevant for system input. I configured conversational conditions and responses, structuring and refining NLP/NLU models for improved contextual understanding. My work included ongoing evaluation of AI performance and redesign of conversational flows to enhance virtual assistant and chatbot accuracy. • Conducted data selection, cleaning, and preprocessing for conversational AI datasets. • Configured and evaluated AI conversational systems using linguistic and business metrics. • Structured and optimized NLP/NLU models to interpret textual data accurately. • Designed conversational flows for virtual assistant/chatbot usability and error reduction.

2022

Assistant AI Trainer at Fluxon Technologies

CrowdsourceTextClassification

In my role as Assistant AI Trainer, I was responsible for synthesizing raw data related to economic and mathematical models and verifying data annotations. I validated the annotated data and presented samples to ensure robust AI training for economic and prediction tasks. I also re-annotated data sets to correct AI misclassification and maintain annotation integrity. • Synthesized and processed raw text data sets for AI model consumption. • Verified and validated economic/mathematical data set annotations for model accuracy. • Presented annotated data to AI programs and checked output correctness. • Re-annotated data as necessary to resolve misclassifications or improve data quality.

In my role as Assistant AI Trainer, I was responsible for synthesizing raw data related to economic and mathematical models and verifying data annotations. I validated the annotated data and presented samples to ensure robust AI training for economic and prediction tasks. I also re-annotated data sets to correct AI misclassification and maintain annotation integrity. • Synthesized and processed raw text data sets for AI model consumption. • Verified and validated economic/mathematical data set annotations for model accuracy. • Presented annotated data to AI programs and checked output correctness. • Re-annotated data as necessary to resolve misclassifications or improve data quality.

2016 - 2018

Education

Work History