I am an MSc Data Science graduate based in the UK, with experience in machine learning, clinical data processing, interpretable AI, and applied LLM/RAG workflows.

My current portfolio focuses on two main areas:

• Clinical and healthcare data science — privacy-preserving data pipelines, gait analysis, time-series processing, and interpretable feature engineering
• Applied AI and LLM systems — retrieval-augmented generation, document-grounded question answering, FastAPI backends, and Streamlit interfaces

For a structured overview of my selected projects, please visit:

Zijie Ren - Data Science & AI Portfolio

A local-only Python package for privacy-preserving clinical gait data ETL and biomechanical feature engineering.

• Modular Python package structure
• Local processing of sensitive clinical gait data
• Walking-zone segmentation
• Biomechanical feature extraction including speed, cadence, stride length, and range of motion
• Clear separation between data processing and modelling workflows

Repository: Gait_Data_Pipeline

An end-to-end Retrieval-Augmented Generation application for document-grounded question answering over user-provided PDF documents.

• PDF ingestion and text preprocessing
• Embedding-based retrieval with FAISS
• FastAPI backend for retrieval and answer generation
• Streamlit frontend for interactive document Q&A
• DeepSeek API integration through an OpenAI-compatible client
• Source citation and retrieved evidence display

Repository: llm-research-intelligence-system

Languages & Data Science: Python, pandas, NumPy, scikit-learn, machine learning, feature engineering, model evaluation

AI & LLM Applications: RAG, embeddings, vector search, prompt engineering, FastAPI, Streamlit, FAISS, DeepSeek API

Healthcare & Research Data: clinical data processing, gait analysis, time-series processing, privacy-preserving local workflows, interpretable AI

Visualisation & Reporting: Matplotlib, data storytelling, stakeholder-focused reporting

• LinkedIn: Zijie Ren
• Portfolio: zijie-ren-portfolio