Basil Kirui

Key Skills

Software

Mindrift

Scale AI

Supervisely

Trilldata Technologies

CVAT

Labelbox

Top Subject Matter

Python and FastAPI Specialist with Advanced Data Streaming Skills"

"Experienced Geocoding Specialist in Spatial Data Analysis and Mapping"

"Expert in NLP-Based Question Answering Systems and Information Retrieval"

Top Data Types

Computer Code Programming

Image

Medical Dicom

Top Task Types

Bounding Box

Computer Programming Coding

Geocoding

Prompt Response Writing SFT

Question Answering

Freelancer Overview

With over five years of experience in data labeling and AI training, I have honed my expertise in creating high-quality datasets essential for developing advanced AI models. My proficiency spans across various domains, including natural language processing (NLP), computer vision, and geospatial data analysis. I have successfully led projects involving multilingual text annotation, image and video labeling, and complex geocoding tasks. My meticulous attention to detail ensures the accuracy and reliability of the datasets, which are critical for the performance of AI applications. A key highlight of my career includes working on AI training data for self-driving cars, where I specialized in labeling complex urban environments and traffic scenarios. Additionally, I have contributed to the development of NLP-based question answering systems, leveraging my skills in text generation and evaluation in both English and Spanish. My technical proficiency in Python, FastAPI, and various data annotation tools, combined with my ability to manage and train labeling teams, sets me apart as a versatile and reliable expert in the field of AI training data.

Labeling Experience

Multilingual Text Classification for Sentiment Analysis

LabelboxTextBounding BoxPolygon

This project involved annotating multilingual text data to classify sentiment and recognize emotions in social media posts. The primary tasks were to identify and label sentiments such as positive, negative, or neutral, and detect specific emotions like joy, anger, sadness, and surprise. The data consisted of social media posts in multiple languages, including English, Spanish, and French. To ensure high-quality annotations, strict guidelines were followed, and regular inter-annotator agreement checks were conducted. The project also included creating a comprehensive labeling schema to capture the nuances of sentiment and emotion across different languages.

This project involved annotating multilingual text data to classify sentiment and recognize emotions in social media posts. The primary tasks were to identify and label sentiments such as positive, negative, or neutral, and detect specific emotions like joy, anger, sadness, and surprise. The data consisted of social media posts in multiple languages, including English, Spanish, and French. To ensure high-quality annotations, strict guidelines were followed, and regular inter-annotator agreement checks were conducted. The project also included creating a comprehensive labeling schema to capture the nuances of sentiment and emotion across different languages.

2023 - 2023

Multilingual Text Classification for Sentiment Analysis

CVATTextBounding Box

This project involved the annotation of multilingual text data to classify sentiment and recognize emotions in social media posts. The primary task was to identify and label sentiments such as positive, negative, or neutral, and detect specific emotions like joy, anger, sadness, and surprise. The data consisted of social media posts in multiple languages, including English, Spanish, and French. To ensure high-quality annotations, we adhered to a strict guideline and conducted regular inter-annotator agreement checks. The project also included the creation of a comprehensive labeling schema to capture the nuances of sentiment and emotion across different languages.

This project involved the annotation of multilingual text data to classify sentiment and recognize emotions in social media posts. The primary task was to identify and label sentiments such as positive, negative, or neutral, and detect specific emotions like joy, anger, sadness, and surprise. The data consisted of social media posts in multiple languages, including English, Spanish, and French. To ensure high-quality annotations, we adhered to a strict guideline and conducted regular inter-annotator agreement checks. The project also included the creation of a comprehensive labeling schema to capture the nuances of sentiment and emotion across different languages.

2023 - 2023

Education

Work History