Hi! I'm Amir Masoud Sefidian
Data Science / Machine Learning
W
elcome to my website! I started programming as a C++ developer for AVR microcontrollers when I was 14. After four years of programming for the high school robotics team, I was motivated to pursue Computer Science. I completed my Master's degree in Computer Engineering in 2017 where I worked in the field of Machine Learning and Data Science especially focused on data preprocessing and missing values imputation techniques. Currently, I am a Data Scientist with a passion for developing innovative algorithms and models that can drive insights and value at Vinted. I regularly share some content that I find useful throughout my learning journey on my Academy/Blog to simplify concepts in Machine Learning, Mathematics, Programming, and related topics. Besides computer science, I enjoy playing the piano and listening to music.
Education
2015-2017M.Sc. in Computer Engineering
- Research Area:
- Data Mining
- Machine Learning
- Data Preprocessing
- Research Area:
2011-2015B.Sc. in Computer Engineering
Publications
- Sefidian, Amir Masoud, and Daneshpour, Negin (2020). "Estimating missing data using novel correlation maximization based methods". Applied Soft Computing, 91, 106249.
- Sefidian, Amir Masoud, and Daneshpour, Negin (2019). "Missing value imputation using a novel grey based fuzzy c-means, mutual information based feature selection, and regression model". Expert Systems with Applications, 115, 68-94.
- Sefidian, Amir Masoud, and Daneshpour, Negin (2018). "Applying regression models on subsets with high correlations for a better numeric missing values imputation". Tabriz Journal of Electrical Engineering (TJEE), 48 (3), 1187-1200.
- Sefidian, Amir Masoud, and Daneshpour, Negin (2017). "Using clustering and a hybrid method to fill the numeric missing values". Iranian Journal of Electrical and Computer Engineering (IJECE), 15(3), 233-242.
Skills & Experience
Programming Languages
Proficient in:
Familiar with:
Databases
Tools & Technologies
Projects
- 2022, BulkBoto3: Python package for fast and parallel transferring a bulk of files to S3 based on boto3!
-
2020-2022, As a Machine Learning Engineer in Eveince (formerly ParticleB):
- Developed "A framework to cut down the runtime of walk-forward optimization in algorithmic trading strategies by leveraging the parallel processing capability on the cloud architecture".
- Contributed to developing a Backtesting Framework that provides various tools for implementing and testing trading strategies such as fetching data, extracting features, training predictive models to generate signals, allocating assets, sizing the bets, executing orders, and evaluating portfolio performance.
- Implemented a parameter optimization service that periodically tunes parameters for algorithmic trading strategies.
- Developed a recommendation system to help FIDIBO (the largest digital platform of Ebooks, Audiobooks, and Podcasts in Iran) users discover personalized and relevant recommendations.
- 2020, AI-powered audio source separation (vocals/instruments) web application.
- 2020, 2D Landmark Detection & Robot Tracking using Graph SLAM (Simultaneous Localization and Mapping).
- 2020, Automatic image captioning service trained on the MS COCO dataset.
- 2020, Facial Keypoint Detection system that takes in any image with faces, recognizes and detects faces, and predicts the locations of 68 distinguishing keypoints on each face using Deep CNNs.
- 2019, A machine learning based system that gives insights about metrics of an organization.
- 2018, Designed and implemented a web-based appointment scheduling, accounting, and management system for a consultation institute to efficiently manage their 40k+ appointments, 9k+ transactions, and 4k+ clients.