Abtin Zandi
Software engineer · AI & mobile development
Abtin
Zandi
I build software where useful AI meets the real world, from mobile experiences to computer vision and practical developer tools.
01 / About
Engineering with a practical point of view.
I’m a software engineer at Uritect with a computer engineering background from Amirkabir University of Technology. My public work spans Python, backend and mobile development, machine learning, and computer vision.
I enjoy moving between research and implementation: testing models, building with developer tools, and turning technical ideas into systems people can actually use. I also write about what I learn along the way.
02 / Selected work
Built, explored, and shared.
Featured research project · team work
Seeing obstacles through a smartphone.
ROD is an assistive vision project from Amirkabir University of Technology. The team’s public dataset brings together 24,326 annotated images across 25 obstacle categories to support real-time detection on a phone. Its documented system combines YOLOv8n, ARCore distance estimation, and audio or vibration feedback.
Explore the datasetComputer vision experiments
Public notebooks and projects exploring OpenCV and YOLO.
View repository ↗Deep learning projects
Hands-on Python notebooks around deep learning and generative models.
View repository ↗Backend & mobile
Public learning projects across Python web frameworks and Kotlin development.
Explore GitHub ↗03 / Writing
Notes from the build.
FLUX 2.0, tested with Replicate and Python
Read article ↗LangChain vs raw SDK: two ways to build an agent
Read article ↗FastAPI for modern APIs and AI services
Read article ↗Thoughts on software, AI, and what works in practice.
Visit Medium ↗04 / Background
Curious across the stack.
Education
Bachelor of Science in Computer Engineering, Amirkabir University of Technology (Tehran Polytechnic).
Areas of interest
Applied AI, computer vision, Python, API development, Android, and the work of connecting models to useful products.
05 / Get in touch
Have an idea worth building?
I’m always interested in thoughtful conversations about software, applied AI, and the problems technology can help solve.