Hongxiang Q.

Key Skills

Software

Other

Top Subject Matter

Conversational AI

Emotional Intelligence

Educational AI Assistant

Top Data Types

Text

Top Task Types

Fine-tuning

Prompt + Response Writing (SFT)

Freelancer Overview

AI Model Fine-tuning and Data Preparation - AI Girlfriend Chat Application. Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Bachelor of Management, Beijing Jiaotong University (2023) and Bachelor of Arts, The College at Rochester (2023). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning and Prompt + Response Writing (SFT).

Labeling Experience

Prompt Generation and Response Annotation - AI Study Assistant for College Students

OtherTextPrompt + Response Writing (SFT)

Generated and organized prompts, questions, and responses to train and evaluate AI assistant responses for course Q&A and study-related topics. The project required writing, annotating, and rating conversation pairs to enhance the model's adaptability and personalization to student needs. Focused on creating realistic academic scenarios with high linguistic diversity. • Authored and annotated diverse academic prompts and answers • Performed evaluation and ranking of generated responses • Collaborated on dataset curation for personalized study help • Ensured variety in question types and student learning scenarios

2024 - 2024

AI Model Fine-tuning and Data Preparation - AI Girlfriend Chat Application

OtherTextFine-tuning

Trained and fine-tuned large language models (LLMs) on open-source textual datasets for an AI-powered chat application. The work involved preparing and cleaning data, curating prompts and responses, and iteratively adjusting datasets for optimal language model performance. Integrated retrieval-augmented generation with personal context to simulate realistic interactions and memory. • Fine-tuned Llama2 models for emotional conversation use case • Prepared and annotated prompt/response pairs for dialogue simulation • Evaluated and adjusted training results to ensure natural replies • Used open-source tools and custom scripts for data handling

2024 - 2024

Education

Work History