IAP
Designed and implemented an event-driven, cloud-agnostic platform (EED) for efficient cloud diagnostics.
I participated in designe and implement an event-driven, cloud-agnostic platform (EED) for efficient cloud diagnostics. This project involved developing microservices using Golang and Rust, establishing gRPC communication between services. I deployed a sharded MongoDB database, designing its schema for tracking execution graphs and events, and implemented Kafka as an event broker to handle event processing. Utilizing Test-Driven Development (TDD) practices, I ensured the robustness and reliability of the microservices. I also participated in deployment a Kubernetes cluster in Azure to manage and autoscale the containerized project, and used Terraform to safely deploy infrastructure components, including microservices, databases, and task runners. Additionally, I implemented serverless architecture with OpenFaaS and Docker for task execution.
RestockAlerts
Developed Restock alert app for Shopify stores, and implemented auto-deploy with Gitlab-CI and Docker-compose on azure Virtual machines.
I participated in development of Shopify apps, including a back-in-stock feature, and implemented auto-deploy pipelines using GitLab-CI and Docker-compose on Azure Virtual Machines. I also participated in web application implementation using the Django web framework to facilitate customer interactions through a specialized panel. The project utilized PostgreSQL and MongoDB for database management, leveraging advanced Django ORM for efficient querying and manipulation. Additionally, I automated tasks using Celery to perform statistical analysis and data monitoring.
Astrolabe
A ranking web application like Alexa, for Iranian android applications.
As a Technical Officer for the Trends product, I managed terabytes of data using Google BigQuery and MongoDB, and processed data with Spark. I deployed and maintained web applications on Google Cloud Platform (GCP) using Ansible. I implemented web applications with the Django web framework and managed a React-based project, enhancing its features and ensuring its maintenance. Utilizing Test-Driven Development (TDD) practices with pytest and unit test, I ensured the robustness and reliability of microservices.
TUBP
Predict internet user behavior in long-term (up to one month) periods using usage patterns and activity types.
This project was my masters thesis, marking the first time I was responsible for designing and maintaining a high-throughput system. I monitored terabytes of data, processed it, and saved it in appropriate databases. I chose MongoDB due to the volume and non-relational nature of the data, and utilized Python for data processing with Redis and Celery, as well as for data analysis with Pandas and NumPy. For deployment, I employed Docker-compose and CI/CD pipelines. Additionally, I implemented machine learning models, including LSTM and CNN, for predictive analysis.