Samuel Ngila

Key Skills

Software

CVAT

iMerit

OneForma

Sama

Remotasks

Toloka

Telus

Internal/Proprietary Tooling

Top Subject Matter

Computer Vision

Robotics Domain Expertise

Egocentric Video

Top Data Types

Video

Image

Text

Document

Top Task Types

Segmentation

Bounding Box

Polygon

Classification

Entity (NER) Classification

Point/Key Point

Polyline

Cuboid

Object Detection

Text Generation

Question Answering

Text Summarization

Evaluation/Rating

Data Collection

Freelancer Overview

Tier 3 Labeller. Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other and Sama. Education includes Bachelor of Science, Rongo University (2018) and Certificate, Samasource Digital Basics (2019). AI-training focus includes data types such as Video and Image and labeling workflows including Segmentation.

Labeling Experience

Tier 3 Labeller

OtherVideoSegmentation

As a Tier 3 Labeller at Atlas Capture LLC, I performed advanced annotation of egocentric video data for household task training in robotics and AI. My responsibilities included creating dense and coarse action labels, quality assurance, and refinement of existing annotations, ensuring high accuracy in continuous hand-object interaction scenarios. I consistently applied the single goal mental model and mastered timestamp correction to train autonomous systems effectively. • Executed granular video labeling using dense/coarse segmentation per client standards. • Reviewed and corrected grammatical, spelling, and clarity errors in annotations while maintaining annotator intent. • Audited and improved segment timestamp accuracy to align with fine-grained action windows. • Maintained rigorous quality assurance by identifying and eliminating common annotation errors.

As a Tier 3 Labeller at Atlas Capture LLC, I performed advanced annotation of egocentric video data for household task training in robotics and AI. My responsibilities included creating dense and coarse action labels, quality assurance, and refinement of existing annotations, ensuring high accuracy in continuous hand-object interaction scenarios. I consistently applied the single goal mental model and mastered timestamp correction to train autonomous systems effectively. • Executed granular video labeling using dense/coarse segmentation per client standards. • Reviewed and corrected grammatical, spelling, and clarity errors in annotations while maintaining annotator intent. • Audited and improved segment timestamp accuracy to align with fine-grained action windows. • Maintained rigorous quality assurance by identifying and eliminating common annotation errors.

2025 - Present

AI Training Data Specialist

SamaImageSegmentation

As an AI Training Data Specialist at Sama, I labeled thousands of images and videos to support computer vision model development for major technology clients. My duties included applying multiple annotation methods such as bounding boxes, polygons, keypoints, and instance segmentation on SamaHub and related tools. I also collaborated closely with QA and contributed to multimodal dataset curation for perception-critical autonomous systems. • Executed high-accuracy annotation using bounding box, polygon, and keypoint techniques in SamaHub. • Performed 3D point cloud annotation, ASR evaluation, and OCR analysis for AI assistant training. • Conducted frame-by-frame video tracking and real-time conversational agent annotation projects. • Utilized proprietary data pipelines leveraging JSON and REST APIs for data submission and review.

As an AI Training Data Specialist at Sama, I labeled thousands of images and videos to support computer vision model development for major technology clients. My duties included applying multiple annotation methods such as bounding boxes, polygons, keypoints, and instance segmentation on SamaHub and related tools. I also collaborated closely with QA and contributed to multimodal dataset curation for perception-critical autonomous systems. • Executed high-accuracy annotation using bounding box, polygon, and keypoint techniques in SamaHub. • Performed 3D point cloud annotation, ASR evaluation, and OCR analysis for AI assistant training. • Conducted frame-by-frame video tracking and real-time conversational agent annotation projects. • Utilized proprietary data pipelines leveraging JSON and REST APIs for data submission and review.

2021 - Present

Education

Work History