For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
F
Faith Chepkoech

Faith Chepkoech

Graduate Research Assistant – VEMI Lab

USA flagMaine, Usa
$50.00/hrExpertOther

Key Skills

Software

Other

Top Subject Matter

Machine learning
visualization
Computer Vision

Top Data Types

ImageImage

Top Task Types

Object DetectionObject Detection
ClassificationClassification

Freelancer Overview

Graduate Research Assistant – VEMI Lab. Brings 8+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Other. Education includes Master of Science, University of Maine (2025) and Bachelor of Science, Maseno University (2019). AI-training focus includes data types such as Image and labeling workflows including Object Detection and Classification.

ExpertEnglish

Labeling Experience

Deepfake Detection for Cybersecurity Applications

OtherImageClassification
I participated in training a deepfake detection model by labeling real and fake facial images within a large-scale dataset. Annotation tasks included classifying images for model input and supporting data balancing through augmentation. Attention to labeling accuracy was essential for improving F1 scores and model reliability. • Classified images as real or deepfake to train MesoNet CNN • Contributed to dataset curation and preparation for balanced classes • Adopted precise image labeling protocols for facial authenticity • Used open-source datasets and internal scripts for annotation

I participated in training a deepfake detection model by labeling real and fake facial images within a large-scale dataset. Annotation tasks included classifying images for model input and supporting data balancing through augmentation. Attention to labeling accuracy was essential for improving F1 scores and model reliability. • Classified images as real or deepfake to train MesoNet CNN • Contributed to dataset curation and preparation for balanced classes • Adopted precise image labeling protocols for facial authenticity • Used open-source datasets and internal scripts for annotation

2025 - 2025

Computer Vision-Assisted Audio Guidance for Blind and Low-Vision Passengers (Capstone)

OtherImageObject Detection
I developed and implemented an object detection pipeline for seat availability and door position identification. As part of this work, I labeled and augmented images used for training YOLO-based vision models. Annotation also supported NLP-driven audio guidance features for accessibility. • Labeled vehicle simulator images to train object detection models • Applied bounding box labels for object localization tasks • Performed manual quality checks and data augmentation on labeled images • Employed open source and internal tools for annotation and pipeline automation

I developed and implemented an object detection pipeline for seat availability and door position identification. As part of this work, I labeled and augmented images used for training YOLO-based vision models. Annotation also supported NLP-driven audio guidance features for accessibility. • Labeled vehicle simulator images to train object detection models • Applied bounding box labels for object localization tasks • Performed manual quality checks and data augmentation on labeled images • Employed open source and internal tools for annotation and pipeline automation

2025 - 2025

Graduate Research Assistant – VEMI Lab

ImageObject Detection
I managed the end-to-end process of labeling and curating multimodal datasets for accessibility-focused assistive systems. Data labeling included handling images and text for computer vision and NLP model training and evaluation. Work included failure analysis and iteration based on labeled data results. • Oversaw dataset curation and manual annotation for computer vision tasks • Labeled images for use in object detection and sentiment analysis models • Collaborated on improving annotation guidelines and quality control • Used internal/proprietary tooling within an academic research environment

I managed the end-to-end process of labeling and curating multimodal datasets for accessibility-focused assistive systems. Data labeling included handling images and text for computer vision and NLP model training and evaluation. Work included failure analysis and iteration based on labeled data results. • Oversaw dataset curation and manual annotation for computer vision tasks • Labeled images for use in object detection and sentiment analysis models • Collaborated on improving annotation guidelines and quality control • Used internal/proprietary tooling within an academic research environment

2024 - 2025

Automated Detection of Photophobia Using Computer Vision

ImageClassification
I curated and annotated eye-region images to train and evaluate a ResNet18-based model for photophobia detection. My labeling work focused on precisely tagging pupil dilation, blinking events, and eye movement signals under various lighting conditions. Accurate labeling was key to model performance in real-time detection scenarios. • Labeled facial images for classification of photophobia indicators • Used labeling guidelines for quality assurance and consistency • Assisted in developing data augmentation strategies for rare event classes • Utilized internal toolsets and Python code for labeling workflow

I curated and annotated eye-region images to train and evaluate a ResNet18-based model for photophobia detection. My labeling work focused on precisely tagging pupil dilation, blinking events, and eye movement signals under various lighting conditions. Accurate labeling was key to model performance in real-time detection scenarios. • Labeled facial images for classification of photophobia indicators • Used labeling guidelines for quality assurance and consistency • Assisted in developing data augmentation strategies for rare event classes • Utilized internal toolsets and Python code for labeling workflow

2024 - 2024

Education

U

University of Maine

Master of Science, Data Science and Engineering

Master of Science
2023 - 2025
M

Maseno University

Bachelor of Science, Applied Statistics with Information Technology

Bachelor of Science
2015 - 2019

Work History

M

Midwest Health Analytics Group

Data Scientist

Maine
2025 - Present
U

University of Maine

Graduate Research Assistant

Orono
2024 - 2025