Projects | Alireza Belal

October 2023 - January 2024

Developed a Python-based automated trading system that executes triangular arbitrage strategies on Binance cryptocurrency exchange.

Implemented algorithms to detect price discrepancies across multiple currency pairs, enabling profitable arbitrage opportunities.
Built a real-time execution system capable of completing trades within milliseconds to capitalize on market inefficiencies.
Created a monitoring dashboard to track performance metrics and trading history.
Implemented risk management strategies to protect against market volatility.
Conducted extensive backtesting and optimization cycles resulting in consistent profitability during testing periods.

Python API Integration Algorithm Design Data Analysis Financial Markets

October 2022 - September 2024

Led research project at Sharif University focused on developing deep learning models to predict interactions between drugs and protein targets.

Designed a hybrid model combining deep learning for feature extraction with gradient boosting algorithms for precise prediction of binding affinity.
Processed complex biological datasets including protein sequences, 3D structures, and chemical properties.
Improved model performance by incorporating machine learning algorithms for regression tasks, achieving a 10% accuracy improvement over existing methods.
Utilized neural networks for automatic feature extraction from drug and protein data.
Published research results demonstrating significant performance advantages over traditional methods.

Python TensorFlow PyTorch Neural Networks Bioinformatics Gradient Boosting

February 2023 - September 2023

Created an innovative NLP model to convert classical Persian poetry into modern styles while preserving contextual meaning and literary value.

Developed a specialized neural network architecture to understand and translate the nuances of Persian poetic forms.
Implemented style transfer techniques to maintain the essence of original poems while transforming language structures.
Built a corpus of paired classical and modern Persian poetry to train the model.
Created an evaluation framework to assess both semantic preservation and stylistic transformation.
Achieved 78% user satisfaction in blind tests comparing model outputs with human translations.

NLP Python Text Generation Style Transfer Persian Language Processing

December 2022 - June 2023

Led development of a comprehensive e-commerce solution for AVLIsland, specializing in importing and selling musical instruments in Iran.

Designed and implemented a multi-vendor marketplace architecture allowing independent sellers to manage inventory and fulfill orders.
Created a robust user management system with dynamic permissions based on product categories.
Implemented SEO optimizations resulting in 40% increase in organic traffic.
Developed custom analytics dashboard for tracking sales, inventory, and customer engagement metrics.
Led a cross-functional team of 6 developers, coordinating work through Agile methodologies.

Laravel PHP MySQL RESTful APIs Microservices Agile

December 2021 - January 2025

Developed an advanced fraud detection system for WiseTrack startup to identify suspicious activities in mobile advertising and user acquisition data.

Designed and implemented a real-time anomaly detection system using machine learning models.
Created a scalable backend infrastructure to process and analyze millions of daily transactions.
Integrated multiple data sources to enrich transaction information for more accurate fraud detection.
Developed an alert system to notify stakeholders of potentially fraudulent activities.
Implemented continuous learning algorithms to adapt to evolving fraud patterns.

April 2020 - August 2020

Developed a computer vision solution to detect whether individuals are wearing face masks in real-time video streams.

Built and trained a Convolutional Neural Network (CNN) to identify the presence and proper usage of face masks.
Implemented data augmentation techniques to improve model generalization across diverse lighting conditions and angles.
Optimized the model for real-time processing on standard hardware.
Created a demo application with webcam integration to showcase the technology.
Achieved 94% accuracy in real-world testing scenarios.

Python Computer Vision Deep Learning CNNs Image Processing Data Augmentation

February 2023 - September 2023

Developed a specialized NLP tool for analyzing sentiment in Persian language texts, enabling better content classification for diverse applications.

Created a custom Persian language preprocessing pipeline to handle language-specific nuances.
Built and trained models using ParsBERT and LSTM architectures to classify text sentiment.
Developed a Persian-specific sentiment lexicon to enhance model accuracy.
Implemented a user-friendly API for integration with content management systems.
Achieved 87% accuracy on diverse Persian text datasets, outperforming general-purpose multilingual models.

NLP Python Sentiment Analysis LSTM Persian NLP Machine Learning

December 2021 - January 2025

Designed and implemented a comprehensive reporting system API for WiseTrack, providing real-time analytics and customizable reports.

Developed a flexible API architecture allowing for customized report generation based on user roles and preferences.
Implemented data aggregation and processing pipelines to transform raw data into actionable insights.
Created an efficient data storage solution by leveraging columnar storage and indexing, reducing analysis time from 10 minutes to under 1 second.
Built interactive dashboards to visualize key performance metrics.
Implemented automated report generation and distribution features.

Laravel PHP RESTful APIs Data Visualization