Patrick G.

Key Skills

Software

Other

Appen

Remotasks

Top Subject Matter

LLM fine-tuning

RLHF evaluation

prompt-response quality (accuracy, tone, safety)

Top Data Types

Image

Text

Document

Top Task Types

RLHF

Bounding Box

Question Answering

Freelancer Overview

Senior Data Annotator / QA Reviewer (Contract Ended) — Outlier AI (Remote). Brings 17+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Other, Appen, and Remotasks. Education includes Bachelor of Science (Honours) in Architecture, Jomo Kenyatta University of Agriculture and Technology (2014) and Diploma in Architecture, Jomo Kenyatta University of Agriculture and Technology (2010). AI-training focus includes data types such as Computer Code, Programming, and Image and labeling workflows including RLHF, Bounding Box, and Entity

Labeling Experience

Image & Spatial Data Annotator — Scale AI (Remote)

OtherImageBounding Box

Performed image annotation for architectural and urban imagery used in downstream model training. Applied bounding boxes, polygon segmentation, and object classification to built-environment scenes. Maintained consistent throughput aligned to project SLAs while supporting QA audits. • Drew bounding boxes for relevant objects in architectural/urban imagery. • Created polygon segmentations for spatially precise labeling. • Labeled objects with domain-informed classification categories. • Ensured low error rates through QA checks and SLA adherence.

2025 - 2026

Senior Data Annotator / QA Reviewer (Contract Ended) — Outlier AI (Remote)

OtherRLHF

Provided RLHF ranking and preference labeling support for large language model fine-tuning workflows. Evaluated model-generated responses for quality signals including accuracy, tone, and safety. Helped structure prompt-response assessment across multiple domain categories. • Performed ranking and preference label assignments for LLM training pipelines. • Reviewed responses for accuracy, tone, and safety compliance. • Conducted prompt-response pair evaluation across domain categories. • Rated outputs to support model iteration and QA consistency.

2024 - 2026

Text Classification & Named Entity Recognition Specialist — Appen (Remote)

AppenText

Labeled text datasets for NLP model training across legal, technical, and general-domain content. Performed named entity tagging alongside sentiment and document categorization annotations. Coordinated with project leads to resolve edge cases and maintain guideline consistency. • Tagged named entities in labeled text corpora. • Assigned sentiment labels to support classification training. • Categorized documents by type for downstream NLP tasks. • Reviewed and corrected edge cases to keep labeling consistent.

2024 - 2025

Multimodal Data Labeler — Remotasks (Remote)

RemotasksTextQuestion Answering

Annotated multimodal datasets combining text and images for vision-language model training. Contributed to caption generation and visual question answering labeling, including image-text alignment checks. Flagged ambiguous or policy-sensitive content for escalation via review channels. • Generated or supported caption generation for image-text pairs. • Labeled visual question answering (VQA) outputs. • Verified image-text alignment for consistency. • Escalated ambiguous or policy-sensitive items through established processes.

2024 - 2024

Education

Work History