Ann S.

Key Skills

Software

Other

CVAT

Top Subject Matter

Autonomous driving / traffic scene understanding

E-commerce search relevance / information retrieval

Sports analytics / football event understanding

Top Data Types

Video

Text

3D Sensor

Image

Document

Top Task Types

Action Recognition

Cuboid

Bounding Box

Freelancer Overview

AI Prompt Specialist | Digital Divide Data – Kenya (2024–2025). Core strengths include Other and CVAT. Education includes Bachelor of Computer Science, Kenyatta University (2022). AI-training focus includes data types such as Video, Text, and 3D Sensor and labeling workflows including Action Recognition, Evaluation, and Rating.

Labeling Experience

Data Labeling Assistant | Execo – Remote (2024–2025)

Other3D SensorCuboid

Annotated freight vehicles (trailers, trucks, yard tractors) using 3D point clouds and multi-camera views for CV training. Maintained QA-verified accuracy while labeling sequences at high frame throughput. Collaborated with QA to resolve edge cases and reduce labeling errors, improving dataset quality for autonomous logistics scenes. • 3D point cloud labeling for vehicle instances • Multi-camera 3D CV training annotation • High-throughput frame labeling for sequence consistency • QA partnership to resolve edge cases and reduce errors

2024 - 2025

Sports Data Annotator | Talent Match Kenya – Remote (2024–2025)

OtherVideoAction Recognition

Annotated football events (passes, shots, interceptions, defenses) with frame-level accuracy for AI sports analytics. Produced high QA-verified accuracy labels to support performance analysis and ML models. Delivered real-time annotations for live and recorded matches and worked with AI engineers to refine definitions and reduce inconsistencies. • Frame-level football event annotation (2K+ events) • QA-verified accuracy for ML-ready datasets • Real-time live and recorded match labeling • Refinement of event definitions with engineering teams

2024 - 2025

Relevance Annotator | Upwork – Remote (2024–2025)

OtherText

Evaluated relevance of product search results for specific e-commerce search queries. Compared listings across multiple sources to judge match quality and category alignment with user intent. Followed strict guidelines while reviewing large volumes of content to produce consistent labels for search training and benchmarking. • Relevance judgment per query and listing • Cross-source comparison for accuracy and match quality • Category/context and intent evaluation • Guideline-based consistency across high-volume tasks

2024 - 2025

AI Prompt Specialist | Digital Divide Data – Kenya (2024–2025)

OtherVideoAction Recognition

Annotated multi-camera driving footage to identify road situations, lane conditions, and vehicle behaviors for autonomous driving datasets. Labeled structured lane-change scenario elements (Situation, Goal, Checks, Decision, Action, Post-Action) and assessed traffic, obstacles, and sensor views for consistency. Delivered high-quality labeled data to support autonomous driving decision-making model improvements. • Lane-change scenario labeling with event elements • Multi-camera video assessment for accurate annotations • QA-oriented review for accuracy and consistency • Dataset contribution for autonomous driving training

2024 - 2025

Data Annotator | Remotasks (Scale.ai) – Remote (2020–2025)

Other3D SensorCuboid

Annotated 10K+ LiDAR frames with 3D bounding boxes along with segmentation and keypoints for autonomous driving datasets. Achieved high QA-validated accuracy and delivered dense-scene labeling at consistent daily throughput. Specialized in multi-class 2D/3D alignment across LiDAR and camera projections and iterated with QA/engineering to improve protocols and usability. • LiDAR 3D box, segmentation, and keypoint labeling • Multi-class 2D/3D alignment across sensor projections • QA validation and protocol refinement with engineering • High-precision labeling in dense urban scenes

2020 - 2025

Show more

Education

Work History