Hire the Best Data Scientists
in Taiwan

AI & Platform Engineer | LLMs | ML | Cloud | ETL | K8S | 🎓 Master’s @ UC Berkeley I am a production-focused AI & Platform Engineer specializing in building scalable, autonomous systems. With a Master's from UC Berkeley and enterprise experience at Volvo Cars, I bridge the gap between cutting-edge AI research and reliable cloud infrastructure. I don't just build models; I architect end-to-end solutions that are robust, secure, and ready for business. ⸻ ⭐ What I bring to your project: ✅ AI Agent Development: Building multi-agent systems, RAG, and custom assistants that automate complex workflows. ✅ LLM Engineering: Fine-tuning and hosting local LLMs (Llama, Qwen) and integrating advanced APIs for specialized tasks. ✅ Cloud Platform Architecture: Orchestrating scalable infrastructure with Kubernetes, Terraform, and Crossplane on Azure/AWS. ✅ Backend & Data Systems: Designing high-performance APIs and ETL pipelines that handle enterprise-scale data. ✅ Reliability & Ops: Implementing CI/CD best practices and self-service infrastructure to ensure smooth deployments. ✅ Clear Communication: Collaborative partner who prioritizes transparency and business value. I look forward to building something impactful with you. Cheers! ⸻ 💻 Tech Stack Languages: Python, SQL, R, C#, .NET, Bash AI/ML: PyTorch, scikit-learn, Transformers, LangGraph, GPT, Gemini, Qwen Data Engineering: Apache Spark, ETL Pipelines, Data Lakes, SQL Server, MongoDB Cloud: AWS (SageMaker, Lambda, S3, EC2), Azure (Databricks, Synapse, Pipelines), Terraform, Kuberenetes, Helm Backend & DevOps: REST APIs, Docker, Git Visualization & BI: Tableau, DOMO, Excel Certifications: AWS Cloud Practitioner, Tableau Desktop Specialist ⸻ 📞 Let’s work together to build something powerful — whether it’s an AI-powered solution, a cloud-based data pipeline, or a full-stack system that brings your data to life. I’m responsive, results-driven, and ready to deliver value fast.

I refactor your brittle, high-latency data workflows and volatile schema mutations to engineer clean, secure, production-grade GenAI platforms that eliminate operational lag and guarantee absolute output accuracy. - Slashed LLM hallucinations by 40% and accelerated processing speeds 10x by overhauling legacy monolithic parsing with a layout-aware multi-agent architecture and semantic chunking. - Permanently eliminated a cluster of 75+ daily data defects and reduced pipeline lag by 12 hours across 500,000+ daily records by deploying serverless, SOLID-principled PySpark ETL workflows. - Accelerated executive decision cycles by two weeks and cut underlying API infrastructure costs by 25% by building real-time, metadata-grounded Streamlit and centralized BI ecosystems. Ready to transition your AI infrastructure from an unpredictable, costly prototype into a high-fidelity, secure production engine? Let's connect to map out your architecture.

🚀 Unlock the potential of your data with tailored LLM and Generative AI solutions. With hands-on experience developing full-stack Generative AI and Retrieval-Augmented Generation (RAG) systems using LangChain, OpenAI, and API, I build scalable and production-ready AI solutions. I also specialize in Python-based constraint solvers, such as Timefold, which integrates optimization logic into backend APIs. My expertise is backed by a solid foundation in Pandas, NumPy, and Scikit-learn, enabling efficient data preprocessing and classic machine learning workflows that complement modern LLM applications. ✅ Financial QA Assistant (RAG System) Designed a full-stack Retrieval-Augmented Generation (RAG) chatbot using OpenAI, LangChain, and Pinecone. The assistant answers financial questions from vector-embedded documents via an interactive Streamlit interface. ✅ AI-Driven Math Problem Generation Created a prompt-driven solution using OpenAI’s LLMs to generate structured math exercises. Focused on educational automation and generative QA pipelines. ✅ Timefold-Based Hospital Scheduler Built a hospital shift scheduling backend using Python, FastAPI, and the Timefold constraint solver. Implemented complex planning constraints (availability, skill, shift groups, contract rules), JSON API interfaces, Dockerized deployments, and cURL/Pytest testing workflows. 📣 Client Review: “Cindy completed work on a complex Python project with advanced business logic. I found her to be very thorough and pleasant to work with.” ✅ Predictive Modeling and Analysis Built machine learning pipelines for wine quality prediction and Titanic survival analysis using scikit-learn. Included full data cleaning, feature engineering, and evaluation workflows. 💡 Struggling with Any of These? - Building custom AI assistants using your internal documents? - Need to optimize complex scheduling or planning problems? - Integrating Generative AI into your tools or workflows? Services I Offer: ✨LLM & RAG Systems: LangChain, Pinecone, OpenAI, Streamlit – tailored AI assistants for document-based Q&A and automation ✨ Constraint Solvers: Optimization logic for scheduling, staffing, and logistics (Timefold, OptaPlanner) ✨ Data Science & ML: Classic modeling (classification, regression), feature engineering, scikit-learn pipelines ✨ Generative AI: Custom GPT-driven tools, prompt engineering, content generation ✨ Data APIs & Visualization: RESTful APIs (FastAPI), dashboards, cURL workflows, Pytest-based test automation 📈 Work With Me, and You Will: - Achieve Efficiency: Automate repetitive decision-making with AI or solver logic - Scale Faster: Build tailored tools that handle complexity — from schedules to insights - Stay Ahead: Deploy cutting-edge Generative AI and optimization techniques in production Let’s bring your data, AI, and optimization goals to life — message me to get started!

I am a statistics graduate student with a robust foundation in mathematics and hands-on experience in machine learning and deep learning. My expertise includes developing AI solutions, building AI agents, and creating LLM applications tailored for financial and economic domains. I specialize in leveraging technologies like Python, SQL, and vector databases to automate data workflows and drive results. A key project involved achieving 76.5% accuracy in style identification, showcasing my ability to merge statistical modeling with AI applications effectively. If you seek a results-oriented collaborator who can transform complex challenges into practical solutions, I am eager to discuss how I can contribute to your project. Let's connect and explore the opportunities ahead.

I’m an academic researcher and editor with experience supporting students, scholars, and professionals in writing, editing, and data analysis. Whether you need help polishing a journal article, analyzing survey results, or improving your thesis, I can help. • Skilled in academic research (humanities and social science), academic writing, editing, and proofreading • Proficient in R, Python, Excel, content and sentiment analysis, statistics and survey experiment • Clear communication and timely delivery are important to me—let’s work together