林星亞

Key Skills

Software

Other

Top Subject Matter

AI Language Model Evaluation and Prompt Engineering

Multimodal AI Image Quality Assessment and Feedback

Top Data Types

Text

Image

Top Task Types

Prompt + Response Writing (SFT)

Freelancer Overview

Independent AI Tools & API Researcher. Brings 10+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Bachelor of Arts, 上海大學 (Shanghai University) (2016). AI-training focus includes data types such as Text and Image and labeling workflows including Prompt + Response Writing (SFT), Evaluation, and Rating.

Labeling Experience

Independent AI Tools & API Researcher

OtherTextPrompt + Response Writing (SFT)

As an Independent AI Tools & API Researcher, I provided extensive prompt engineering and output evaluation for advanced language models. My tasks included crafting prompts, analyzing generation results, and iteratively refining prompts to align with accuracy, creativity, and domain-specific tasks. I routinely tested and compared the effectiveness of different models using various APIs for research purposes. • Deep experience with Claude Pro, Google AI Pro, and Gemini API for LLM evaluation • Designed, reviewed, and refined text prompts to improve generation quality • Conducted backend testing, parameter tuning (Temperature, etc.), and SFT-style prompt+response creation • Evaluated model outputs for factual accuracy, coherence, and stylistic alignment

2023 - Present

Freelance Photographer & Content Creator – AI Visual Evaluator

OtherImage

In my freelance practice as a professional photographer and subject matter expert, I offered expert visual analysis and review for imagery intended for multimodal AI training. My work involved detailed evaluation of artistic and technical image attributes, identifying issues, and giving targeted feedback to improve dataset quality. I assessed both photographic and generated images to ensure they met the required standards for AI development. • Used Capture One and Profoto along with digital and analog cameras for image creation and review • Evaluated image datasets for quality, style, artistic merit, and technical accuracy • Provided structured feedback to support RLHF and other human-in-the-loop AI training workflows • Supported error analysis, style diversity, and model tuning for better multimodal results

2019 - Present

Education

Work History