How AI Can Help Urban Planners to Create City Concepts
Arthur Kazaryan. (2021). "How AI Can Help Urban Planners to Create City Concepts." Geek Culture.
CV
Education
- M.Sc., ITMO University, St. Petersburg (online). 2024
- Professional retraining in Data Science, University of Artificial Intelligence, Moscow, Russia (online). 2021
- B.Arch. in Urban Planning, Saint Petersburg State University of Architecture and Civil Engineering, St. Petersburg, Russia. 2018
Work experience
Data Scientist / Python Backend. Apr ‘22 - . Metaperson.
The project is about creating a video with lip-synchronized animations for different purposes: localization, marketing, presentation, etc. As a software developer I achieved the following goals:
- took part in making a pipeline for creating a realistic lip-synchronized videos;
- created an API service for DL pipeline based on FastAPI, Celery, Redis and Flower wrapped in a docker-compose;
- created a Django web-service with Celery task manager for creating lip-synchronized videos and text-to-speech audiofiles via web interface;
- created a Telegram Bot based on the same Django web-service.
Junior Data Scientist / Python Backend. Apr ‘21 - Mar ‘22. University of Artificial Intelligence.
The project was created to allow users to build a neural networks based on Keras framework using web interface. I have done:
- algorithms for file processing and continuous support of old ones;
- algorithms for creating datasets for a number of tasks: classification, segmentation, object detection, etc;
- a layout of a web interface. Worked in a team with designers and front-end developers for implementation;
- and published a documentation for dataset creation process on readthedocs.org.
Data Science Intern. Dec ‘20 - Mar ‘21. University of Artificial Intelligence.
A newly created project for building neural networks using web interface. The achievements at the end of internship were:
- wrote an algorithms for processing raw data such as images, videos, texts, audio, table data;
- wrote a module for creating and loading datasets and using it in the process of training neural network.
Skills
- Programming
- Python, Git, Scikit-learn, Keras, OpenCV, NumPy, Pandas, Librosa, Matplotlib, Seaborn, FastAPI, Django, Celery.
- Software
- PyCharm, Visual Studio Code, Anaconda, Jupyter.