Alireza Zamanian

AI Engineer & Consultant · PhD in Computer Science

Profile

AI engineer with a PhD in Computer Science from TUM and almost ten years of experience across data science, applied AI research and production AI software, much of it in healthcare. Currently leading an AI development team at SYNLAB, one of Europe's largest medical diagnostics providers.

Experience

AI Development Team Leader

Jan 2026 – today

AI Product Engineer

Nov 2024 – Jan 2026

SYNLAB International, Munich

Software & AI Consultant

2026 (4 months)

PoriZma

Doctoral Researcher

Sep 2021 – Nov 2024

Fraunhofer IKS, Munich

Doctoral Researcher

Sep 2019 – Aug 2021

Helmholtz Munich, Munich

Education

PhD in Computer Science

2019 – 2026

Technical University of Munich (TUM)

Magna cum laude. Thesis: Achieving Trustworthiness in Medical AI via Domain-Informed Analysis

M.Sc. Robotics Engineering

2015 – 2017

Amirkabir University of Technology (Tehran Polytechnic)

B.Sc. Mechanical Engineering

2009 – 2014

Amirkabir University of Technology (Tehran Polytechnic)

Skills

AI & machine learning
LLMs, retrieval-augmented generation (RAG), AI agents, deep learning, time series, explainable AI
Statistics & data
Statistical modeling, missing data, causal inference, sensitivity analysis, study design
Engineering
Software development, Python, Microsoft Azure, production AI systems
Domains
Healthcare and medical diagnostics, robotics

Languages

English
Fluent (working language)
Persian
Native
German
Intermediate (B1)

Publications

  1. Evaluation of Active Feature Acquisition Methods for Time-varying Feature Settings (opens in a new tab) H. von Kleist, A. Zamanian, I. Shpitser, N. Ahmidi Journal of Machine Learning Research 26(60) · 2025
  2. Analysis of Missingness Scenarios for Observational Health Data (opens in a new tab) A. Zamanian, H. von Kleist, O. A. Ciora, M. Piperno, G. Lancho, N. Ahmidi Journal of Personalized Medicine 14(5) · 2024
  3. KI-Entwicklung: von der Vorschrift zum Computercode A. Zamanian, E. Pachl, G. Lancho Medizintechnik · 2024
  4. Assessable and Interpretable Sensitivity Analysis in the Pattern Graph Framework for Nonignorable Missingness Mechanisms (opens in a new tab) A. Zamanian, N. Ahmidi, M. Drton Statistics in Medicine 42(29) · 2023
  5. What about the Latent Space? The Need for Latent Feature Saliency Detection in Deep Time Series Classification (opens in a new tab) M. Schröder, A. Zamanian, N. Ahmidi Machine Learning and Knowledge Extraction 5(2) · 2023
  6. Post-hoc Saliency Methods Fail to Capture Latent Feature Importance in Time Series Data (opens in a new tab) M. Schröder, A. Zamanian, N. Ahmidi Trustworthy Machine Learning for Healthcare (TML4H), Springer LNCS · 2023
  7. Evaluation of Active Feature Acquisition Methods for Static Feature Settings (opens in a new tab) H. von Kleist, A. Zamanian, I. Shpitser, N. Ahmidi arXiv:2312.03619 · 2023
  8. Partially Specified Causal Simulations (opens in a new tab) A. Zamanian, L. Mareis, N. Ahmidi arXiv:2309.10514 · 2023
  9. Early-, Late-, and Very Late-Term Prediction of Target Lesion Failure in Coronary Artery Stent Patients: An International Multi-Site Study (opens in a new tab) E. Pachl, A. Zamanian, M. Stieler, C. Bahr, N. Ahmidi Applied Sciences 11(15) · 2021
  10. Estimation of Moving Obstacle Dynamics with Mobile RGB-D Camera (opens in a new tab) S. B. Germi, A. Zamanian, M. A. Arzati, M. A. Khosravi, R. FesharakiFard International Conference on Robotics and Mechatronics (ICRoM) · 2017