Frank Oigara

Key Skills

Software

Labelbox

Label Studio

Remotasks

Top Subject Matter

AI Code Evaluation and LLM Training

Document Annotation for Recruitment/OCR Pipelines

LLM Evaluation

Top Data Types

Document

Text

Video

Top Task Types

Entity (NER) Classification

Transcription

Text Summarization

Object Detection

Question Answering

Freelancer Overview

Independent Software Engineer & AI Trainer. Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Labelbox and Label Studio. Education includes Bachelor of Science, Kenyatta University (2020). AI-training focus includes data types such as Computer Code, Programming, and Document and labeling workflows including Computer Programming, Coding, and Entity (NER) Classification.

Labeling Experience

AI Trainer/LLM Evaluation Specialist

Label StudioTextRLHF

Participated in reinforcement learning from human feedback (RLHF) as part of LLM evaluation cycles. Evaluated, ranked, and audited LLM outputs and responses for quality, helpfulness, bias, and hallucinations. Performed red teaming and process reward annotation tasks for model safety and robustness. • Created and reviewed preference pairs and DPO training data • Executed red teaming and hallucination audits for generated text • Applied multi-step reasoning and prompt engineering • Used Label Studio, Labelbox, and proprietary pipelines for annotation workflows.

Participated in reinforcement learning from human feedback (RLHF) as part of LLM evaluation cycles. Evaluated, ranked, and audited LLM outputs and responses for quality, helpfulness, bias, and hallucinations. Performed red teaming and process reward annotation tasks for model safety and robustness. • Created and reviewed preference pairs and DPO training data • Executed red teaming and hallucination audits for generated text • Applied multi-step reasoning and prompt engineering • Used Label Studio, Labelbox, and proprietary pipelines for annotation workflows.

2021 - Present

Data Labeling Specialist (RecGlobal Project)

LabelboxDocumentEntity Ner Classification

Produced structured, schema-consistent document annotations across bilingual (EN/AR) recruitment and OCR pipelines. Applied annotation guidelines for document structure, entity extraction, and data validation in production web applications. Strong focus on document annotation for admin dashboards and automated workflows in EN/AR environments. • Utilized Labelbox and Label Studio in RecGlobal recruitment portal • Worked with passport scans and admin workflow documents • Emphasized guideline compliance and labeling consistency • Annotations supported multilingual AI model training and validation.

Produced structured, schema-consistent document annotations across bilingual (EN/AR) recruitment and OCR pipelines. Applied annotation guidelines for document structure, entity extraction, and data validation in production web applications. Strong focus on document annotation for admin dashboards and automated workflows in EN/AR environments. • Utilized Labelbox and Label Studio in RecGlobal recruitment portal • Worked with passport scans and admin workflow documents • Emphasized guideline compliance and labeling consistency • Annotations supported multilingual AI model training and validation.

2021 - Present

Independent Software Engineer & AI Trainer

Labelbox

Reviewed and annotated AI-generated code in multiple programming languages for correctness and best practices. Evaluated code for logical soundness, efficiency, and edge case coverage as part of OpenTrain-style workflows. Produced and refined code data for supervised fine-tuning (SFT) of large language models and supported RLHF and preference ranking tasks. • Languages handled: JavaScript, Python, Kotlin, and Java • Involved in code generation review, step verification, and chain-of-thought processes • Used Labelbox, Label Studio, and GitHub workflows for annotation and QA • Covering tasks such as RLHF preference ranking, reasoning step audits, and code-based SFT annotation.

Reviewed and annotated AI-generated code in multiple programming languages for correctness and best practices. Evaluated code for logical soundness, efficiency, and edge case coverage as part of OpenTrain-style workflows. Produced and refined code data for supervised fine-tuning (SFT) of large language models and supported RLHF and preference ranking tasks. • Languages handled: JavaScript, Python, Kotlin, and Java • Involved in code generation review, step verification, and chain-of-thought processes • Used Labelbox, Label Studio, and GitHub workflows for annotation and QA • Covering tasks such as RLHF preference ranking, reasoning step audits, and code-based SFT annotation.

2021 - Present

Education

Work History