Joe

Key Skills

Software

Other

Top Subject Matter

Medical AI (multimodal consultation, RAG in healthcare domain)

Enterprise AI workflows (LLM chat + knowledge base in low-code environment)

AI workflow construction (visual orchestration of LLM/RAG pipelines)

Top Data Types

Text

Document

Computer Code Programming

Top Task Types

Question Answering

Function Calling

Data Collection

Freelancer Overview

Orthopedic Multimodal AI Q&A System (Node.js Backend). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. Education includes Bachelor of Science, Qingdao University of Technology (2023) and Associate Degree, Zibo Vocational Institute (2020). AI-training focus includes data types such as Text, Document, and Computer Code and labeling workflows including Question Answering, Function Calling, and Prompt + Response Writing (SFT).

Labeling Experience

AI Workflow Visual Editor Platform

Prompt + Response Writing (SFT)

Built a visual AI workflow editor platform using open-source workflow solutions as a base. Implemented and expanded customized features by fixing upstream bugs from the community repositories to meet internal business operation needs. Delivered an environment used for constructing AI workflows, enabling repeatable prompt and response generation steps within the editor. • Workflow editor built on concepts from n8n and Flowise. • Upstream bug fixes and feature expansion for internal requirements. • Enabled visual orchestration of AI pipeline steps for consistent outputs. • Improved reliability of AI workflow construction and execution for teams.

2024 - 2026

Enterprise Low-code Platform Customization

DocumentFunction Calling

Customized an enterprise low-code platform by embedding AI workflow engines and providing AI chat and knowledge base modules for internal usage. Added an AI chat widget and a knowledge base module to enable document preview and AI-assisted knowledge retrieval workflows. Extended the platform to support custom global pages so internal teams could deploy branded AI-enabled entry points. • Embedded Flowise AI workflows and n8n process engines into the low-code platform. • Built knowledge base module supporting document preview and chat-based access. • Implemented platform extensions enabling custom global pages for deployment. • Resolved upstream issues and adapted tooling into a proprietary internal product.

2024 - 2026

Orthopedic Multimodal AI Q&A System (Node.js Backend)

TextQuestion Answering

Developed and led a multimodal medical AI Q&A backend that answers user consultations using an LLM integrated with a medical RAG repository. Implemented retrieval to ground responses in relevant medical knowledge and generated diagnostic advice plus recommended reference materials. Built Human-in-the-loop oriented AI workflow components as part of enterprise medical consultation pipelines. • Data involved multimodal user inputs (image + text) and retrieval-augmented generation outputs. • LLM integration for consulting features and recommendation of medical videos/materials. • Server-side implementation using NestJS for the mini-program backend. • Emphasis on medical domain relevance via RAG repository connections and generated outputs.

2024 - 2026

Frontend Development Engineer (2D SCADA + Dashboard Designer)

Contributed to the development of AI-related frontend systems that support upstream AI components and interactive dashboards. Delivered a 2D SCADA graphic editor and a visual dashboard designer that connects backend component configurations to canvas injections. While not direct annotation work, this enabled operational visualization and interactive interfaces commonly used to support AI system monitoring and evaluation. • Backend-driven component configuration wired into frontend canvas rendering. • Interactive dashboards supporting real-time status visualization for downstream AI systems. • Implemented history management and event bindings enabling reliable assessment of system behavior. • Optimized rendering to prevent freezes under high-density graph workloads.

2022 - 2024

Offline Data Synchronization Proxy

Data Collection

Implemented offline data synchronization and resilient client-side storage to ensure continuous capture of user inputs/forms during network outages. Built a proxy layer using Axios interceptors to store data locally and retry queued execution when connectivity returns. This improved data collection reliability for systems that depend on user-submitted information. • Interceptor-based local persistence of user inputs/forms. • Silent queue retry after reconnection to ensure data continuity. • Robust offline handling across unreliable factory floor networks. • Supported reliable downstream processing that can feed AI/analytics pipelines.

2020 - 2022

Education

Work History