Allen B.

Key Skills

Software

No software listed

Top Subject Matter

Healthcare workflow assistance

medical triage and appointment routing

Programming education

Top Data Types

Text

Top Task Types

Fine-tuning

Freelancer Overview

Large-Model Medical Consultation & Appointment Assistant | Team Lead (AI training relevance: LLM recommendation evaluati. Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include RAGFlow and PyTorch Transformer. Education includes Bachelor of Engineering, Wuhan College (2026). AI-training focus includes data types such as Medical, DICOM, and Text and labeling workflows including Evaluation, Rating, and Fine-tuning.

Labeling Experience

Large-Model Medical Consultation & Appointment Assistant | Team Lead (AI training relevance: LLM recommendation evaluation and knowledge-base curation)

Led an LLM-based medical consultation and appointment assistant with an emphasis on evaluating recommendation outputs against user symptom context. Validated and refined a RAGFlow medical knowledge base by documenting edge cases and improving the quality of retrieved content used for downstream guidance. Performed QA-style output validation to ensure the system’s recommendations aligned with symptoms and routing needs. • Evaluated model recommendations using symptom-context matching • Validated RAG knowledge-base content quality and completeness • Documented edge cases and failure modes for iteration • Performed workflow QA via functional and API-oriented testing of the product back end

2024 - 2025

PI-Model Teaching Service Assistant Platform | Project Member (AI training relevance: text/code data structuring, classification, and model fine-tuning)

TextFine-tuning

Developed an NLP-based conversational programming-learning platform and created large volumes of structured programming knowledge records via cleaning, classification, and categorization. Fine-tuned a Transformer model to improve intent recognition and answer quality in programming QA scenarios, then assessed inference performance. Used evaluation-oriented iteration with a focus on improving classification logic and response quality for user dialogue. • Cleaned and categorized web-crawled programming knowledge into structured records • Built classification and intent-recognition support for conversational QA • Fine-tuned a Transformer model for answer quality improvement • Validated model outputs in programming-learning dialogue scenarios

2022 - 2023

Education

Work History