Machine Learning Specialization
DeepLearning.AI, Stanford University, Prof. Andrew Ng · Oct 2025
A Developer who wants to be a regular, everyday, normal ML Engineer.
Current Position
Master's student in AI at Ca' Foscari University of Venice
Graduation Date
Expected June 2027
Field of interest
Machine Learning · Computer Vision
About
Before starting my Master's degree, I worked as a Backend Developer for 4 years, and took a bachelor's degree in Computer Science in my beautiful country, Iran. I am currently a second-year student and am seeking an opportunity to work or research in the field of Machine Learning and Computer Vision.
Education
-
Ca' Foscari University of Venice
Master’s Degree in Artificial Intelligence and Data Engineering
Sep 2024 – exp. June 2027
Course CFU Grade Foundations of Artificial Intelligence and Machine Learning 12 27/30 Geometric and 3D Computer Vision 6 30/30L Image and Video Understanding 6 27/30 Software Architectures 6 25/30 Algorithms and Learning over Massive Data 12 22/30 Calculus and Optimization 6 23/30 Information Retrieval and Web Search 6 24/30 Cryptography 6 Spring 2026 Applied Probability for Computer Science 6 Spring 2026 Cloud Computing and Distributed Systems 6 Spring 2026 Advanced Data Management 6 Spring 2026 Statistical Inference and Learning 6 Spring 2026 Deep Learning for Natural Language Processing 6 Fall 2026 Final thesis 24 Fall 2026 Internship 6 Fall 2026 -
University of Bojnord
Bachelor’s Degree in Computer Science
Sep 2012 – Jun 2016
Course Grade Fundamentals of Computer and Programming 18.5/20 Algorithms and Data Structures 20/20 Advanced Programming 16.52/20 Differential Equations 20/20 Fundamentals of Combinations 20/20 General Mathematics 20/20 Computer Graphics 19/20 Principles of Operating Systems 20/20 Theory of Calculation 18/20 Topics in Computers Sciences 20/20 Principles of Software Design 20/20 Bachelor Project of Computer Sciences 20/20
Certificates
TOEFL iBT
ETS · Mar 2023
101/120
Articles
Machine Learning Approaches for Imbalanced Intracranial Pressure Analysis 'Submitted to special session 6: Learning and Intelligent Optimization for Digital Healthcare Systems'
LION20 · Feb 2026
I am working on finding a better approach to classify a highly correlated, imbalanced dataset to predict the intracranial pressure (ICP) signals of patients.Currently, I achieved good results using different classical, tree-based and deep learning models, feature engineering, and data augmentation techniques.The goal is to reduce the false positive rate to the minimum possible, while keeping the balance in Specificity and Sensitivity and of course, high AUC score.
A novel method for solving universum twin bounded support vector machine in the primal space
Springer · Nov 2023
In this article we propose (NUTBSVM), a Newton-based approach for solving in the primal space the optimization problems related to Twin Bounded Support Vector Machines with Universum data (UTBSVM). In the NUTBSVM, the constrained programming problems of UTBSVM are converted into unconstrained optimization problems, and a generalization of Newton’s method for solving the unconstrained problems is introduced.
Projects
Ketamine ICP
LION20 · Feb 2026
Here are some results for the ICP dataset using different approaches. Missing values were handled with KNN imputation, feature selection for classical models was performed using PCA (95% variance explained), and class imbalance was addressed using a combination of SASMOTE and class-weighted sampling.
Model comparison (train/test metrics; test confusion matrix)
| # | Model | Tr AUC μ | Tr AUC σ | Tr t (s) | Te AUC | Sens | Spec | TP | TN | FP | FN | Te t (s) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | CNN-BiLSTM-Attention | 0.9992 | 0.0001 | 2237 | 0.9787 | 0.9696 | 0.8951 | 4626 | 128 | 15 | 145 | 3433 |
| 2 | Dilated-CNN | 0.9621 | 0.0009 | 96 | 0.9141 | 0.8405 | 0.8462 | 4010 | 121 | 22 | 761 | 182 |
| 3 | CatBoost | 0.9882 | 0.0009 | 1 | 0.9317 | 0.9382 | 0.8112 | 4476 | 116 | 27 | 295 | 2 |
| 1D-CNN (Base) | 0.9307 | 0.0106 | 75 | 0.8738 | 0.8132 | 0.7762 | 3880 | 111 | 32 | 891 | 143 | |
| XGBoost | 0.9845 | 0.0012 | 1 | 0.9203 | 0.9401 | 0.7692 | 4485 | 110 | 33 | 286 | 1 | |
| Attention-CNN | 0.9519 | 0.0043 | 188 | 0.8926 | 0.8742 | 0.7622 | 4171 | 109 | 34 | 600 | 630 | |
| Extra Trees | 0.9880 | 0.0011 | 1 | 0.9221 | 0.9554 | 0.7343 | 4558 | 105 | 38 | 213 | 1 | |
| LightGBM | 0.9983 | 0.0002 | 5 | 0.9467 | 0.9805 | 0.6993 | 4678 | 100 | 43 | 93 | 5 | |
| Random Forest | 0.9982 | 0.0003 | 12 | 0.9416 | 0.9782 | 0.6923 | 4667 | 99 | 44 | 104 | 15 | |
| CNN-LSTM | 0.9913 | 0.0005 | 495 | 0.9280 | 0.9602 | 0.6853 | 4581 | 98 | 45 | 190 | 947 | |
| ResNet-1D | 0.9759 | 0.0019 | 84 | 0.9019 | 0.9447 | 0.6643 | 4507 | 95 | 48 | 264 | 157 | |
| MLP | 0.9973 | 0.0002 | 49 | 0.9107 | 0.9799 | 0.6573 | 4675 | 94 | 49 | 96 | 105 |
3D Laser Scanner
3D Geometry and Computer Vision Course · Mar 2026
Using a calibrated camera, 2 reference planes with known distance, and a laser line, we can calculate the depth of each pixel in the image. We can then use this depth information to create a 3D point cloud of the scene.
A website for studying better powered by AI
Productivity project · Oct 2025
Term
Active recall
Definition
Self-testing beats passive rereading; the app uses spaced review the same way.
Anki is a perfect tool for studying, but a web based version powered by AI could be more convenient.You can upload different files formats and use your own AI agent to summarize the content.You can also share your decks with others and use them on your mobile or desktop.
RePAIR Project
Introduction to Machine Learning · Sep 2024
The goal of this project was to go deeper into the source code of YOLOv8 and modify it to improve the performance of the model,represented in the article that was published by the RePAIR project. Results showed a higher mAP50 score in instance segmentation and bounding boxes.
Customized YOLOv8l — validation metrics: bounding boxes (left) and instance segmentation (right).
| Model | Box P | Box R | Box mAP50 | Seg P | Seg R | Seg mAP50 |
|---|---|---|---|---|---|---|
| Customized YOLOv8l | 0.7866 | 0.8659 | 0.8439 | 0.8961 | 0.8113 | 0.9025 |
Article Reading Automation
Productivity project · Oct 2025
Automation using n8n, Google Gemini, Postgres, and a FastAPI site to make daily article reading more productive.
Semi-Supervised SVM vs Newton Universum Twin SVM
Introduction to Artificial Intelligence · Apr 2023
Implemented Semi-Supervised SVM (S3VM) and Newton-based Universum Twin SVM (Newton-UTSVM); compared performance and proposed Unconstrained S3VM combining both approaches.
Benchmark comparison: accuracy % (top) and runtime in seconds (bottom); bold = best in row for that metric.
| Dataset (samples × features) | Newton | S3VM constrained | S3VM unconstrained |
|---|---|---|---|
| Diabetes Pima (768 × 8) | 76.630.03 |
75.4813.50 |
75.100.49 |
| Ionosphere (350 × 34) | 86.550.03 |
86.5512.35 |
84.031.40 |
| Musk (476 × 166) | 86.420.25 |
81.48158.12 |
80.863.12 |
| Breast Cancer (568 × 31) | 97.410.03 |
97.9358.16 |
94.304.18 |
| Sonar (207 × 60) | 85.710.02 |
80.006.43 |
67.140.90 |
| Gender (5001 × 7) | 96.470.80 |
96.299993.28 |
95.7122.28 |
Multi-Class Normally Distributed Cluster Centers Data Generator
Jul 2020 · with Hossein Moosaei & Dave Musicant
NDC generates random centers for multivariate normal distributions, separating planes, and class labels; measures separability by points on the wrong side of the plane. Integer-valued for simplicity.
MC-NDCC generator · 3D scatter preview · CSV download
Loading chart…
Dental Assistant Project
Python, Django, PyTorch · Feb 2023
Panel for dentists to manage patient records and X-rays; AI model flags decayed or at-risk teeth for clinical review.
Work experience
Backend Developer
Aug 2022 – Mar 2024
Respina Network & Beyond, Tehran, Iran
Python
Django
Flask
RESTful APIs
Celery
RabbitMQ
PostgreSQL
Docker
Kubernetes
Git
CI/CD (GitLab CI)
Prometheus
Loki
Grafana
Linux
Nginx
Troubleshooting & Performance Optimization
Asterisk
ARI
Respina is a leading provider of telecommunications solutions in Iran, offering dedicated internet access, SIP-Trunk, Hosted-PBX phone services, and data center colocation. Trusted for reliability and innovation, Respina enables seamless connectivity and robust communication for businesses, contributing to Iran's digital infrastructure and economic growth. This was one of the most impactful periods of my career, where I gained deep hands-on experience in analyzing, testing, and optimizing large-scale backend systems. As part of the Hosted-PBX (Nexfon) team, I worked on improving the performance, scalability, and reliability of enterprise-grade telecommunication services. Redesigned and optimized the billing system using CGRATES, implementing a real-time charging solution that improved billing accuracy and efficiency. Reduced monthly reporting time from 30 minutes to 40 seconds by optimizing database queries, parallelizing tasks with Celery, and improving data aggregation workflows. Refactored the Asterisk-ARI integration into an event-driven Flask microservice, containerized with Docker, and enhanced with multi-processing, increasing concurrent call capacity per instance from 25 to 130 and reducing infrastructure load by nearly 5×. Implemented comprehensive monitoring and troubleshooting using Prometheus, Loki, and Grafana, improving observability and system reliability. Collaborated with the DevOps team to migrate deployments to a Kubernetes-based cloud infrastructure, enhancing scalability, fault tolerance, and CI/CD automation.
Backend Developer
Nov 2021 – Jul 2022
ANIL Web design studio, Tehran, Iran
PHP
Laravel
JavaScript
CSS3
HTML5
RESTful APIs
Postman
MySQL
Git
Docker
Linux
Nginx
Agile Collaboration
Collaborated with a highly skilled development team composed of top university talents, gaining valuable exposure to modern software design practices and collaborative workflows. Contributed to two major web projects — RadmanPack and PelikanIran — by developing and integrating backend systems using PHP and the Laravel Framework, while coordinating closely with frontend developers working in React.js. Focused on designing RESTful APIs, implementing secure authentication, and ensuring smooth data exchange between backend and frontend components. Additionally, gained hands-on experience in version control (Git), containerization with Docker, and deployment on Linux-based environments, emphasizing maintainability, scalability, and team collaboration.
Backend Developer
Oct 2020 – Jul 2021
AvinAvisa Lab, Tehran, Iran
Node.js
Express.js
MongoDB
JavaScript
Blockchain Integration (Ethereum
TRON)
RESTful APIs
Git
Docker
Linux
WebSocket Communication
Worked in a research-driven environment within the Blockchain Laboratory of Amirkabir University of Technology, focusing on the development of decentralized systems and blockchain-based applications. Contributed to the design and implementation of a cryptocurrency exchange platform (Polychain) enabling peer-to-peer trading with advanced matching logic based on trade volume, customer tier (VIP levels), and other dynamic factors. Implemented a custom matching algorithm inspired by the Knapsack problem, optimizing trade pair selection and transaction efficiency. Additionally, integrated the platform with Ethereum and TRON networks, handling blockchain interactions and ensuring secure, real-time transaction processing. Collaborated with a small, multidisciplinary team, applying the MERN stack (MongoDB, Express.js, React.js, Node.js) and blockchain APIs to build a robust, scalable backend system.
Military Service
Jul 2018 – Jul 2020
Army, Tehran, Iran
In Iran, military service is compulsory for men and must be completed in order to obtain permission to leave the country. The service lasts for two years. During this period, I collaborated with Dr. Moosaei (Charles University) and Dr. David Musicant (Carleton College) on one publication and one research project, while also engaging in self-study in machine learning and image processing to enhance my technical expertise.
Android Developer
Aug 2016 – Apr 2018
Ishaya, Tehran, Iran
Java
Android SDK
RESTful APIs
MySQL
Git
XML
Material Design
Started my professional journey in software development as an Android Developer, initially joining as an intern and later continuing as a full-time engineer due to strong performance and enthusiasm for mobile technologies. Contributed to the development of Ponila, the first intelligent content recommendation system for Persian-language users. The platform utilized semantic and syntactic analysis of Persian text to deliver personalized article recommendations based on user interests, improving content discovery and engagement. Collaborated with backend and data science teams to integrate recommendation algorithms and ensure seamless data synchronization between the mobile app and server-side APIs. Focused on building a responsive, stable, and scalable Android application aligned with modern UI/UX principles.
Contact
Email
me72atgmail.com
Phone
+393516158597