Akason J.

Key Skills

Software

Scale AI

Appen

Remotasks

Top Subject Matter

LLM fine-tuning datasets for RLHF/SFT and safety alignment

NLP classification and entity extraction for enterprise clients

General AI training data including image segmentation and medical/radiology annotations

Top Data Types

Text

Image

Top Task Types

RLHF

Entity (NER) Classification

Segmentation

Freelancer Overview

Senior Data Annotation Specialist (Lead annotator) — RLHF and SFT annotation for LLM fine-tuning. Core strengths include Scale AI, Appen, and Remotasks. Education includes Bachelor of Science, University of Houston–Downtown (2018). AI-training focus includes data types such as Text, Medical, and DICOM and labeling workflows including RLHF, Entity (NER) Classification, and Segmentation.

Labeling Experience

Senior Data Annotation Specialist (Lead annotator) — RLHF and SFT annotation for LLM fine-tuning

Scale AITextRLHF

Lead annotator for RLHF and SFT data used in large language model fine-tuning, including instruction following, summarization, code generation, and safety alignment tasks. Authored and revised annotation guidelines across three product verticals to improve consistency and accelerate onboarding for new annotators. Conducted systematic quality audits on weekly batches and escalated ambiguous edge cases and policy conflicts to project stakeholders. • Labeled instruction-following and summarization outputs for model alignment • Evaluated code correctness and performed safety alignment judgments • Performed QA audits on 500–1,000 items per week with reliability checks • Escalated policy conflicts and documented edge-case guidance for training

2022 - Present

Data Labeling Analyst — NLP/NLU annotation for sentiment, NER, intent classification (and object detection)

AppenTextEntity (NER) Classification

Contributed to multiple client annotation projects spanning sentiment analysis, named entity recognition (NER), intent classification, and object detection. Met throughput targets of 400+ annotation units per hour for text classification while maintaining over 95% accuracy on gold-standard items. Participated in calibration and inter-rater reliability reviews to resolve labeling disagreements and refine guidance for edge cases. • Labeled text for sentiment, intent classification, and entity spans • Supported object detection annotation work for Fortune 500 clients • Completed bilingual annotation in English and Twi for dataset expansion • Performed QA review contributions as a Tier 2 annotator

2020 - 2022

Freelance Data Annotator — multi-platform AI data annotation (segmentation, transcription, relevance, conversation quality)

RemotasksSegmentation

Completed multi-platform data annotation tasks including image segmentation, transcription, search relevance rating, and AI conversation quality evaluation. Maintained top-rated performance with a 4.9/5.0 quality score across 40,000+ completed tasks. Annotated medical imaging data, including radiology scan descriptions, following strict compliance protocols. • Labeled image regions via segmentation tasks • Transcribed and rated search relevance and conversation quality • Annotated medical imaging content under HIPAA-compliant procedures • Delivered high-volume, high-quality annotations across varying task types

2019 - 2020

Education

Work History