Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m an AI developer focused on building practical LLM-powered applications, including AI agents, RAG knowledge assistants, and intelligent document processing systems. I help turn business requirements into working solutions, from initial design and development to deployment and ongoing improvement. During my internship at Hello, I developed an internal HR AI agent that integrated enterprise knowledge retrieval, tool calling, data analysis, and automated reporting. At LG Sciencepark, I built a document processing pipeline using PaddleOCR and LLMs, achieving 90% extraction accuracy for key business fields. My technical toolkit includes Python, FastAPI, LangChain, LangGraph, and vector databases such as Milvus, Qdrant, and Chroma. I’m also pursuing a master’s degree in Big Data and AI Analytics at Sejong University in South Korea. What sets me apart is my ability to connect AI engineering with real business needs. I work closely with stakeholders, pay attention to data quality and error handling, and build applications that are maintainable and useful in everyday workflows.…I’m an AI developer focused on building practical LLM-powered applications, including AI agents, RAG knowledge assistants, and intelligent document processing systems. I help turn business requirements into working solutions, from initial design and development to deployment and ongoing improvement. During my internship at Hello, I developed an internal HR AI agent that integrated enterprise knowledge retrieval, tool calling, data analysis, and automated reporting. At LG Sciencepark, I built a document processing pipeline using PaddleOCR and LLMs, achieving 90% extraction accuracy for key business fields. My technical toolkit includes Python, FastAPI, LangChain, LangGraph, and vector databases such as Milvus, Qdrant, and Chroma. I’m also pursuing a master’s degree in Big Data and AI Analytics at Sejong University in South Korea. What sets me apart is my ability to connect AI engineering with real business needs. I work closely with stakeholders, pay attention to data quality and error handling, and build applications that are maintainable and useful in everyday workflows.

YANGHANG





I’m an AI developer focused on building practical LLM-powered applications, including AI agents, RAG knowledge assistants, and intelligent document processing systems. I help turn business requirements into working solutions, from initial design and development to deployment and ongoing improvement. During my internship at Hello, I developed an internal HR AI agent that integrated enterprise knowledge retrieval, tool calling, data analysis, and automated reporting. At LG Sciencepark, I built a document processing pipeline using PaddleOCR and LLMs, achieving 90% extraction accuracy for key business fields. My technical toolkit includes Python, FastAPI, LangChain, LangGraph, and vector databases such as Milvus, Qdrant, and Chroma. I’m also pursuing a master’s degree in Big Data and AI Analytics at Sejong University in South Korea. What sets me apart is my ability to connect AI engineering with real business needs. I work closely with stakeholders, pay attention to data quality and error handling, and build applications that are maintainable and useful in everyday workflows.…I’m an AI developer focused on building practical LLM-powered applications, including AI agents, RAG knowledge assistants, and intelligent document processing systems. I help turn business requirements into working solutions, from initial design and development to deployment and ongoing improvement. During my internship at Hello, I developed an internal HR AI agent that integrated enterprise knowledge retrieval, tool calling, data analysis, and automated reporting. At LG Sciencepark, I built a document processing pipeline using PaddleOCR and LLMs, achieving 90% extraction accuracy for key business fields. My technical toolkit includes Python, FastAPI, LangChain, LangGraph, and vector databases such as Milvus, Qdrant, and Chroma. I’m also pursuing a master’s degree in Big Data and AI Analytics at Sejong University in South Korea. What sets me apart is my ability to connect AI engineering with real business needs. I work closely with stakeholders, pay attention to data quality and error handling, and build applications that are maintainable and useful in everyday workflows.

Available to hire

I’m an AI developer focused on building practical LLM-powered applications, including AI agents, RAG knowledge assistants, and intelligent document processing systems. I help turn business requirements into working solutions, from initial design and development to deployment and ongoing improvement.
During my internship at Hello, I developed an internal HR AI agent that integrated enterprise knowledge retrieval, tool calling, data analysis, and automated reporting. At LG Sciencepark, I built a document processing pipeline using PaddleOCR and LLMs, achieving 90% extraction accuracy for key business fields.
My technical toolkit includes Python, FastAPI, LangChain, LangGraph, and vector databases such as Milvus, Qdrant, and Chroma. I’m also pursuing a master’s degree in Big Data and AI Analytics at Sejong University in South Korea.
What sets me apart is my ability to connect AI engineering with real business needs. I work closely with stakeholders, pay attention to data quality and error handling, and build applications that are maintainable and useful in everyday workflows.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Beginner

Language

Korean

Fluent

English

Advanced

Work Experience

AI Application Intern at Shanghai Hello Puhui Technology Co., Ltd.

June 1, 2026 - September 1, 2026

AI Intern (Artificial Intelligence) at LG Science Park ch.

March 1, 2025 - August 1, 2025

Built an HR AI agent from requirement gathering through solution design, full-stack development, deployment, and iteration. Designed an LLM + RAG agent loop with intent recognition, task decomposition, knowledge retrieval, tool selection/execution, analytics, and report generation. Implemented a RAG pipeline for HR policies and business documents (parsing, chunking, embedding, retrieval) and refined prompting, context construction, and retrieval strategies to improve relevance and consistency. Exposed HR metric calculations, knowledge retrieval, analytics, and report generation as callable tools; integrated with databases/knowledge bases/internal services via APIs/tool calling. Implemented an end-to-end reporting workflow combining structured data, retrieved knowledge, and analytical findings into dashboards and AI-generated reports, collaborating with HR stakeholders on testing, release validation, and evaluation. Developed a document AI pipeline using PDF preprocessing, PaddleOCR, LL

AI Customer Service Agent for Financial Lending (Intern/Project)

January 1, 2025 - Present

Used LLMs for semantic correction, contextual interpretation, and field extraction; applied prompt design, field mapping, and predefined JSON schemas to handle OCR errors, missing characters, misaligned fields, and inconsistent terminology. Implemented field-level checks for completeness, data types, formatting, and business rules with logging, retrieval, and manual fallback for OCR/model failures and malformed JSON. Automated conversion of unstructured business documents into validated structured data for downstream analytics, system integration, and business process automation. Built a lending-focused customer service system covering products from multiple banks, including bank statement and credit report parsing, loan product matching, and domain-specific Q&A. Developed the agent using LangChain and locally deployed Qwen 3-8B, and built a Milvus knowledge base with banking product rules, FAQs, and business workflows achieving ~95% retrieval success rate. Generated embeddings with Qw