Abednego Andati

Key Skills

Software

Appen

Mindrift

Clickworker

Remotasks

Toloka

Top Subject Matter

LLM Fine-tuning

AI-generated Text

Computer Vision

Top Data Types

Text

Image

Audio

Top Task Types

Fine-tuning

Object Detection

Bounding Box

Text Generation

Transcription

Evaluation/Rating

Prompt + Response Writing (SFT)

Freelancer Overview

Data Annotator – AI Training. Brings 1+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include RWS annotation suite, Appen, and Mindrift. Education includes Bachelor of Business Administration, University of Embu (2023) and Certificate in Computer Applications, Kaveye Vocational Training Centre (2017). AI-training focus includes data types such as Text and Image and labeling workflows including Fine-tuning, Object Detection, and Evaluation.

Labeling Experience

Data Annotator – AI Training

TextFine Tuning

As a Data Annotator for LLM fine-tuning at RWS, I processed and evaluated text passages for AI training. My work focused on ensuring quality and guideline adherence in model improvement tasks. I maintained high consistency and accuracy across a quota-driven annotation workflow. • Processed 500+ text items weekly for model fine-tuning and evaluation. • Reviewed AI-generated summaries, identifying hallucinations and factual inaccuracies. • Participated in calibration sessions for guideline updates and ambiguity resolution. • Consistently exceeded throughput quotas while maintaining top quality metrics.

As a Data Annotator for LLM fine-tuning at RWS, I processed and evaluated text passages for AI training. My work focused on ensuring quality and guideline adherence in model improvement tasks. I maintained high consistency and accuracy across a quota-driven annotation workflow. • Processed 500+ text items weekly for model fine-tuning and evaluation. • Reviewed AI-generated summaries, identifying hallucinations and factual inaccuracies. • Participated in calibration sessions for guideline updates and ambiguity resolution. • Consistently exceeded throughput quotas while maintaining top quality metrics.

2024 - Present

Text Annotation Specialist

MindriftText

As a Text Annotation Specialist with Mindrift, I evaluated LLM responses for quality and adherence to guidelines. I re-annotated disputed datasets and documented ambiguous cases for team learning. My mentorship extended to junior team members working on logical reasoning tasks. • Ranked LLM outputs for helpfulness, honesty, and harmlessness. • Re-annotated 2,000 examples, improving inter-annotator agreement from 82% to 91%. • Identified and documented over 50 edge cases for guideline inclusion. • Coached junior annotators on complex annotation assignments.

As a Text Annotation Specialist with Mindrift, I evaluated LLM responses for quality and adherence to guidelines. I re-annotated disputed datasets and documented ambiguous cases for team learning. My mentorship extended to junior team members working on logical reasoning tasks. • Ranked LLM outputs for helpfulness, honesty, and harmlessness. • Re-annotated 2,000 examples, improving inter-annotator agreement from 82% to 91%. • Identified and documented over 50 edge cases for guideline inclusion. • Coached junior annotators on complex annotation assignments.

2024 - 2025

AI Data Annotator

AppenImageObject Detection

As an AI Data Annotator with Appen Crowdgen, I labeled image and video datasets for computer vision applications. Tasks included object detection, attribute classification, and structured feedback for model improvement. I also undertook audio transcription projects to support diverse AI training pipelines. • Labeled objects and attributes across 20+ classes in images and videos. • Maintained a 92% inter-annotator agreement among a distributed annotation team. • Provided feedback reports on annotation results and model failure cases. • Completed more than 1,200 accurate audio transcription tasks.

As an AI Data Annotator with Appen Crowdgen, I labeled image and video datasets for computer vision applications. Tasks included object detection, attribute classification, and structured feedback for model improvement. I also undertook audio transcription projects to support diverse AI training pipelines. • Labeled objects and attributes across 20+ classes in images and videos. • Maintained a 92% inter-annotator agreement among a distributed annotation team. • Provided feedback reports on annotation results and model failure cases. • Completed more than 1,200 accurate audio transcription tasks.

2020 - 2025

Education

Work History