Jas L.

Key Skills

Software

Data Annotation Tech

Appen

Scale AI

Surge AI

Datasaur

Tagtog

Labelbox

Other

Top Subject Matter

Urban Planning & Geographic Data

Architecture & Infrastructure Design

Regulatory Compliance & Technical Writing

Top Data Types

Text

Document

Geospatial Tiled Imagery

Top Task Types

Prompt + Response Writing (SFT)

RLHF

Evaluation/Rating

Question Answering

Text Summarization

Text Generation

Red Teaming

Fine-tuning

Freelancer Overview

Independent AI Evaluator & Prompt Engineer (2024–Present). Brings 11+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Urban Planning Logic and Subject Matter Expertise. AI-training focus includes data types such as Text and Document and labeling workflows including Evaluation, Rating, and Prompt + Response Writing (SFT).

Labeling Experience

Independent AI Evaluator & Prompt Engineer (2024–Present)

OtherText

Conducted side-by-side benchmarking evaluations of multiple LLMs on professional architectural and legal question answering. Applied RLHF-style assessment criteria (Helpfulness, Honesty, Harmlessness) and identified logical inconsistencies and hallucination risks. Produced evaluation findings and refinements to guide model improvements and prompt changes. • Benchmarking across GPT-4, Claude 3.5, and Gemini • Rating and scoring of response quality using HHH • Detecting domain-specific AI hallucinations and logic errors • Feeding detailed feedback to drive iterative refinement

2024 - Present

SME: Urban Planning Data Verification (2016–2024)

OtherDocumentPrompt + Response Writing (SFT)

Processed and transformed large volumes of unstructured regulatory documents into structured, high-logic summaries suitable for supervised fine-tuning. Performed data integrity assurance by running final logic checks on government reports to verify consistency among spatial data, policy text, and technical parameters. Curated and structured internal design guidelines into AI-consumable formats to support downstream technical auditing and automated checks. • Millions of words of regulatory-code summarization for domain knowledge • Logic-checking 50+ government reports for alignment across modalities • Converting design guidelines into AI-readable formats • Ensuring structured reasoning outputs for training-ready datasets

2016 - 2024

Education

Work History