Geoffrey O.

Key Skills

Software

CVAT

Datumbox

Google Cloud Vertex AI

Img Lab

Label Studio

Remotasks

Roboflow

Surge AI

V7 Labs

Appen

AWS SageMaker

Top Subject Matter

Autonomous Vehicles (Self-driving systems)

Medicine / Healthcare-edical diagnosis evaluation, clinical reasoning, treatment recommendations, patient communication, medical image interpretation, pharmacology, oncology.

Engineering-Technical design review, systems engineering, troubleshooting, optimization problems, safety & compliance checks.

Top Data Types

Image

Text

Audio

Top Task Types

Segmentation

Bounding Box

Polygon

Point/Key Point

Entity (NER) Classification

Text Generation

Fine-tuning

Text Summarization

Object Detection

Red Teaming

Transcription

Evaluation/Rating

RLHF

Computer Programming/Coding

Data Collection

Question Answering

Cuboid

Freelancer Overview

Youth Digital Support Professional. Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Education includes Bachelor of Science, South Eastern Kenya University (2023).

Labeling Experience

Traffic Object Labeling Project

ImageSegmentation

Data Collection Thousands of road images are collected from CCTV cameras, dashcams, or public datasets. Labeling Process Using tools like CVAT or LabelImg, annotators: Draw bounding boxes around objects in each image Label each object as: Car Bus Motorcycle Pedestrian Example: A car in an image is boxed and tagged “car” A person walking is boxed and tagged “pedestrian” Quality Check Another reviewer checks: If all objects are labeled If labels are correct and consistent Dataset Preparation The labeled images are converted into formats like YOLO or COCO and split into: Training set Validation set Test set Outcome The final dataset is used to train an object detection model that can: Detect cars and people in real-time video Draw boxes around them with confidence scoresGoing through images one by one, marking important objects, and giving them names so a machine learning model can learn to recognize them automatically

2025 - Present

Image Annotation and Object Detection

ImagePoint/Key Point

An Image Annotation and Object Detection project involves preparing and labeling images so that a machine learning model can learn to identify and locate objects automatically. It starts with collecting and cleaning image data, then annotating objects using tools like bounding boxes or polygons to label items such as people, vehicles, or animals. The labeled data is reviewed for accuracy and formatted into datasets suitable for training. A computer vision model (e.g., YOLO or Faster R-CNN) is then trained to recognize these objects and predict their locations in new images. The model is evaluated using performance metrics like accuracy and mAP, improved through tuning, and finally deployed in real-world applications such as surveillance, agriculture, or automation systems.

2022 - 2022

Education

Work History