Alex M.

Key Skills

Software

Label Studio

CloudFactory

Remotasks

Micro1

Top Subject Matter

Hand movements

object interactions

temporal video annotation quality review

Top Data Types

Video

Audio

Text

Image

3D Sensor

Document

Top Task Types

Action Recognition

Segmentation

Transcription

Object Detection

Point/Key Point

Freelancer Overview

Video Data Reviewer | Micro1 (Remote). Core strengths include Micro1, Label Studio, and Labeling Tool. Education includes Bachelor of Science, Kenyatta University (2022). AI-training focus includes data types such as Video, Audio, and Text and labeling workflows including Action Recognition, Segmentation, and Transcription.

Labeling Experience

Video & Image Annotator | Turing (Remote)

Label StudioVideoSegmentation

Labeled key visual and narrative components in video and image data to support large multimodal AI systems. Annotated temporal sequences and scene transitions using only visually confirmable evidence while maintaining high accuracy. Performed episodic memory annotation by selecting precise time intervals with grounded second-person responses based on first-person recordings. • Labeled tone, lighting, spatial relationships, and motion cues • Validated AI-generated outputs for invalid questions and speculation • Conducted Stage 1 short-segment labeling and chunking of tasks • Performed quality checks for PII exposure, segmentation errors, and overlay defects

2026 - Present

Video Data Reviewer | Micro1 (Remote)

Micro1VideoAction Recognition

Performed high-precision review of model-generated video annotations for accuracy and precision in describing hand movements and object interactions. Verified action timing by ensuring labeled activity boundaries aligned exactly to start and end moments. Assessed caption granularity to confirm each label represented a single independent action. • Corrected action timing and boundaries • Validated left vs right hand usage • Flagged low-quality, ambiguous, or incomplete labels for rework • Provided actionable feedback for model improvement

2026 - Present

Transcription Specialist | RWS (Remote)

AudioTranscription

Refined machine-generated audio segmentation and transcriptions for language AI speech-recognition datasets. Ensured phonetic accuracy, correct punctuation, and compliance with RWS project-specific style guides. Transcribed diverse audio domains including conversational speech, broadcast media, and field recordings while maintaining accuracy rates above 97%. • Corrected ASR errors including mispronunciations and dialect variations • Addressed speaker diarization and segmentation misalignment issues • Resolved ambiguous segments with QA reviewers using asynchronous collaboration • Delivered transcription and segmentation data aligned to evolving annotation schemas

2025 - 2026

AI Data Specialist | Sigma AI (Remote)

Text

Labeled and evaluated large volumes of text, audio, and conversational AI outputs to support training and fine-tuning of NLP and large language models. Assessed response quality for accuracy, coherence, factual correctness, and safety compliance using granular ratings and written justifications. Performed structured labeling including intent classification, entity recognition, sentiment tagging, and instruction-following evaluation across multilingual datasets. • Applied Sigma annotation frameworks and rubrics for consistent decisions • Conducted red-teaming/adversarial testing by identifying policy violations and model failure modes • Maintained high inter-annotator agreement and met daily targets remotely • Focused on Swahili and English language dataset work

2025 - 2025

Data Analyst | CloudFactory (Remote)

CloudFactoryImageObject Detection

Annotated large-scale image datasets using 2D bounding boxes, polygon annotations, and semantic segmentation to support computer vision model training. Applied keypoint annotation for anatomical landmarks and object reference points to enable spatial learning and body pose estimation. Performed image tagging and content classification across thousands of assets, and reviewed annotations to flag errors and guideline deviations. • Processed video annotation tasks with frame-by-frame labeling and object tracking • Organized datasets by category, quality, and relevance • Maintained accuracy and productivity targets for advanced platform tasks • Ensured high dataset integrity through quality checks and agreement monitoring

2024 - 2025

Show more

Education

Work History