Bahar Azari

About Bahar Azari

I am a Research Scientist at Bosch Center for AI (BCAI), where I focus on multimodal generative modeling, vision-language systems, and diffusion-based models. My recent work includes developing CLIP-guided diffusion models and autoencoder-based semantic controls for image manipulation and industrial anomaly detection. My projects at Bosch have led to several patents and reflect my interest in creating interpretable and controllable AI systems that address real-world needs.My technical skills include Python, PyTorch, JAX, deep learning frameworks, representation learning, and applied mathematical modeling. I am also experienced with large-scale data curation, model deployment, and working with high-performance compute clusters.Before joining BCAI, I completed my PhD in Electrical and Computer Engineering at Northeastern University. My doctoral research centered on hierarchical probabilistic modeling, deep state-space models, and the use of mathematical tools—such as group theory and Fourier analysis—to design neural networks that can incorporate domain knowledge and data symmetries. This work led to several publications in top academic venues.I am passionate about translating research into impactful applications and am always open to connecting with others in the AI community.

Experience

1. Machine Learning Research Scientist Ii

   Bosch Center For Artificial Intelligence Bcai

   Jan 2024 — Present

Education

• Amirkabir University of Technology - Tehran Polytechnic

  Bachelor of Science - BS, Electrical, Electronics and Communications Engineering

  2007 — 2011

• Politecnico di Milano

  Master of Science - MS, Information Technology, Telecommunication

  2012 — 2014

• Northeastern University

  Doctor of Philosophy - PhD