Chia En Chang (張嘉恩)
I am an
Specializing in Large Language Models (LLMs), Computer Vision, and Full-Stack Solutions
About Me
As a Computer Science graduate from the University of Arizona, I am an AI Engineer with a strong passion for software development and machine learning. I have hands-on experience in Large Language Model (LLM) development, AI system integration, and advanced image analysis automation. I excel at transforming innovative ideas into effective, AI-driven solutions, from building intelligent systems that integrate language models and computer vision to developing full-stack interactive platforms for complex data analysis and reporting challenges.
Work Experience
AI Research & Technology Engineer
Anivance AI
- Built LLM applications with GPT-4o, Vertex AI, and Claude for medical image captioning, domain-specific Q&A, and automated report writing.
- Enabled AI-driven automation of CellProfiler pipelines for batch cell image processing, labeling, and morphological feature extraction.
- Developed a full-stack interactive platform with Flask and Gradio, simplifying the AI report generation process.
- Implemented an MCP architecture for automatic registration and cross-calling among AI models, enhancing modularity and scalability.
- Researching and developing a Multimodal Neural-Symbolic AI architecture integrating Logic Tensor Networks (LTN) with knowledge graphs, replacing traditional fine-tuning with axiom-driven learning to achieve explainable reasoning processes.
WordPress Developer
Taiwan Residential and Commercial Real Estate
- Developed and launched a comprehensive social platform for the real estate sector using WordPress.
- Managed full-stack development, ensuring seamless front-end and back-end integration to enhance community engagement.
Adversarial Machine Learning Research Assistant
University of Arizona Artificial Intelligence Laboratory
- Conducted research on adversarial machine learning for cybersecurity applications, focusing on evading commercial malware detectors.
- Explored Universal Adversarial Malware Perturbations (UAMP) to improve evasion rates for malware detection systems.
Bachelor of Science in Computer Science
University of Arizona
- Majored in Computer Science with a focus on algorithms, AI problem solving, machine learning, and natural language processing.
Technical Skills
AI/ML Frameworks & Tools
LLM (Claude, OpenAI API, Gemini), RAG, Multi-Agent Systems, Neural-Symbolic AI, LTN (Logic Tensor Networks), FastAPI, Flask
- 大型語言模型 (LLM)：OpenAI GPT-4/GPT-5, Anthropic Claude (Sonnet 4.5), Google Gemini, Vertex AI
- 多模態AI：BLIP, LLaVA, ImageBind, BioCLIP, CLIP
- 神經符號AI：Logic Tensor Networks (LTN), 神經符號混合架構
- 嵌入模型：OpenAI Embeddings, Microsoft Embeddings, 自定義嵌入模型
- NLP：spaCy, scispacy, transformers, DeBERTa
- 框架：FastAPI, Flask, PyTorch
Computer Vision & Image Processing
CellPose, EasyOCR, Pillow, Scikit-Image, OpenCV, PyTorch
- 圖像分析：OpenCV, scikit-image, PIL/Pillow
- 細胞圖像分析：CellProfiler, Cellpose, 生物標記檢測
- 圖像分割：語義分割、實例分割
- OCR：EasyOCR, 文字識別與提取
- 深度學習框架：PyTorch, TensorFlow
Programming Languages
Python (Proficient), Java, JavaScript, PHP, C/C++, SQL
- Python：精通，用於 AI/ML 開發、數據處理、後端開發
- Java：物件導向程式設計、企業級應用開發
- JavaScript：前端開發、互動式網頁應用
- PHP：後端網頁開發、WordPress 開發
- C/C++：系統程式設計、效能優化
- SQL：資料庫查詢與管理
Backend & Web Development
RESTful APIs, WebSocket, Uvicorn, Gradio, HTML, CSS, JavaScript
- Web 框架：FastAPI, Flask
- API 設計：RESTful API, WebSocket, Streaming Response
- 異步處理：asyncio, 背景任務處理
- 前端技術：HTML5, CSS3, JavaScript (Vanilla), Chart.js
- 伺服器：Uvicorn, Gunicorn
- UI 框架：Gradio, Streamlit
Cloud & DevOps
Google Cloud AI Platform (Vertex AI), API Authentication, Ngrok, Git, VS Code
- Google Cloud Platform：Vertex AI, GCP 認證
- 容器化：Docker, docker-compose
- 部署：Uvicorn, ngrok
- 版本控制：Git, GitHub, GitLab
- 開發工具：VS Code, PyCharm
- API 認證：OAuth, JWT, API Key 管理
Other Proficiencies
Neo4j (Knowledge Graph), NLP (NLTK), Prompt Engineering, Python-Docx, Asyncio, MCP Protocol
- 圖資料庫：Neo4j
- 向量資料庫：FAISS, Elasticsearch, MongoDB Atlas
- 知識圖譜：Neo4j 知識圖譜構建與查詢
- 文檔處理：python-docx, pdfplumber, 自動生成與格式化
- 數據處理：pandas, numpy
- 數據可視化：matplotlib, seaborn, Chart.js
- Prompt Engineering：LLM 提示詞優化與設計
- MCP Protocol：模型通訊協議
Key Projects
AnivanceVerse - 虛擬實驗室系統
即時 LLM 實驗員表現評估系統，類似 nof1.ai。各 LLM 扮演實驗員，依照 protocol 在虛擬實驗室進行實驗，實驗結果與真人數據比較，評估各 LLM 實驗員的準確性。
WRG_API - AI驅動週報生成系統
由 AI 驅動的強大後端服務，自動化實驗數據分析與報告生成流程。能夠處理 Word 文件、實驗圖片（如細胞影像）以及數據表格（CSV/Excel），並根據使用者指令，智能地選用分析工具、執行運算，最終產出一份結構化的實驗報告。
ClinicalPaperGenerator - 智能臨床論文生成工具
使用 AI 智能分析醫學文檔和圖片，自動生成高品質學術論文。
Textara - 學術AI報告生成系統
基於多模型AI的智能學術報告生成系統，支援國科會計畫書、期刊論文、學位論文等多種學術文件格式。
CellPainting - 細胞圖像分析系統
整合 CellProfiler 進行細胞圖像分析，包含自動化管線生成、特徵提取與數據分析。
NeuroSymbolicAI - 神經符號AI五層架構系統
實現可解釋、可成長的生物醫學 AI 系統，採用神經符號混合架構（NeSy + LTN），取代傳統 Fine-tuning 方法。
WeeklyReportGenerator - 週報生成器
自動化實驗週報生成系統，整合多種圖像分析工具，生成結構化的實驗報告。
Contact
I am actively seeking full-time opportunities in software development. I look forward to contributing my skills to your team. Please feel free to get in touch!