Description

I help clients solve complex data and AI problems with practical, reliable, and scientifically grounded solutions. My background combines machine learning, computational neuroscience, medical AI, statistical modelling, high-dimensional data analysis, and interpretable AI.

I stand out by combining strong engineering skills with research-level thinking. I do not just train models. I help define the right modelling strategy, design robust validation, avoid data leakage, build clean Python pipelines, and make results interpretable and trustworthy.

My strongest areas are healthcare AI, neuroimaging, brain injury, multimodal clinical data, predictive modelling, anomaly detection, uncertainty quantification, generative AI, model interpretation, and end-to-end Python data science workflows.

Industry field of expertise

Languages

Persian
Native or bilingual
English
Native or bilingual
German
Conversational

Workplace preferences

Remote only

Primarily works remotely

University of Cambridge
Research Associate
January 2025 - Today (1 year and 5 months)
Cambridge, UK
Developed Generative AI models to synthesize normative (healthy) data for patient comparison and severity estimation using multimodal neuroimaging data.

Applied XAI methods to quantify the relevance of clinical variables to outcomes in patients with brain injuries.

Developed interpretable multimodal AI models combining MRI graphs, CT imaging, and blood biomarkers to predict patient outcomes.

Supervised and mentored PhD students on research projects involving machine learning and neuroimaging analysis.
artificial intelligence Machine learning
LMU Munich
Scientific Researcher
January 2020 - January 2025 (5 years)
Munich, Germany
• • Applied ML, RL, and Bayesian models to multimodal datasets to predict cognitive processes in the human brain.
• • Applied LLMs to predict joint cognitive decisions from conversational text data.
• • Managed and optimized online and in-lab experimental frameworks, including EEG setups, to improve data collection quality and reliability.
Golestan Energy Trading Company
ML Researcher
January 2019 - January 2020 (1 year)
Golestan, Iran
• • Created a multi-feature electrical load dataset with geographic and environmental factors like temper ature, wind speed, and humidity. series