Shaista Shabbir

Industrial AI Researcher · TU Dortmund University · Lamarr Institute

✉️ Email GitHub LinkedIn

Education

M.Sc. Computer Science — AI & Data Science
TU Dortmund University, Germany
Focus: Explainable AI, Privacy in NLP, Industrial Machine Learning. Thesis: XAI and Fairness in Machining Process Monitoring.
2022–2024
B.Sc. Computer Science
Pakistan
Data Structures, Algorithms, Database Systems, Software Engineering.
2017–2021

Research Experience

Research Associate — Industrial AI
TU Dortmund University · Lamarr Institute for Machine Learning
Applied XAI (SHAP, LIME, Anchors) to chatter detection in CNC machining. Developed membership inference attack framework for NLP models trained on industrial sensor data. Built RAG and knowledge graph systems for real-time disaster intelligence. Open-source contributions: ResearchOS (AI peer review platform), Navigate Germany (multilingual immigrant guide).
2024–present

Technical Skills

AI / ML
PyTorch · TensorFlow · scikit-learn · LangChain · SHAP · Optuna · FAISS · ChromaDB
Signal Processing
NumPy · SciPy · FFT · FRF Analysis · Acoustic Sensors · CNC Monitoring
Backend / Data
FastAPI · PostgreSQL · ETL · Docker · MLflow · GitHub Actions
Frontend
Next.js 14 · React · TypeScript · Tailwind CSS · HTML/CSS/JS

Selected Projects

ResearchOS
AI peer review quality platform · FastAPI · Next.js · Claude API
Multi-turn chat, reproducibility badge system, claim verification, reviewer fatigue analysis. Live: research-os-phi.vercel.app
2025
Navigate Germany
Multilingual immigrant guide · 10 languages · GitHub Pages
14 module pages, PWA, salary calculator, 6 journey roadmaps, community tips, Claude-powered chat. Used by international students and workers.
2025
Disaster Event Intelligence Platform
USGS · GDACS · ReliefWeb · LangGraph · Knowledge Graph
Real-time multi-source disaster ingestion, GraphRAG-style KG reasoning, 93% severity classification accuracy, 0% hallucination rate.
2026

Languages

🇵🇰 Urdu (native) 🌍 English (C1) 🇩🇪 German (B2)