I'm a passionate Machine Learning Researcher and Distributed Systems Engineer focused on creating scalable, high-performance AI solutions. My work bridges the gap between cutting-edge ML algorithms and production-ready distributed architectures.
-
Building Semi-Supervised Decision Trees with the Semi-CART Algorithm (2024)
International Journal of Machine Learning and Cybernetics -
Enhancing Classification with Semi-Supervised Deep Learning Using Distance-Based Sample Weights (2025) Oral Presentation on International Conference on Machine Learning Technologies and publish in IEEE Xplor, Preprint Arxiv
-
SemiCART - Semi-supervised CART implementation with weighted GINI-Index
- Achieved 12% accuracy improvements on UCI benchmarks
- Implemented novel weighting mechanisms for unlabeled data
-
SemiDeep - SemiDeep helps your models focus on the right data, so they generalize better, especially when labels are scarce, noisy, or imbalanced.
- Achieved +9% on UCI datasets for classifications
- Implemented novel weighting mechanisms for unlabeled data
- Enabling weighted training on Neural Networks
-
Rust Decision Tree - High-performance decision tree implementation
- Memory-efficient design for large datasets
Languages:
- Go
- Python
- Rust
- PHP
- SQL
Machine Learning:
- Semi-supervised learning
- Decision trees
- PyTorch & TensorFlow
- scikit-learn & XGBoost
Distributed Systems:
- Microservices
- Kubernetes
- Kafka & NATS
- Redis
- High-throughput systems
Infrastructure:
- Docker
- K8s
- AWS
- Terraform
- Prometheus & Grafana- MSc in Software Engineering | IAU, South Tehran Branch (2018)
Thesis: Optimizing the XGBoost Algorithm for Semi-Supervised Datasets - BSc in Software Engineering | IAU, Kashan Branch (2013)
- Semi-supervised learning algorithms for data-efficient ML
- Distributed ML systems with high reliability and performance
- Uncertainty-aware algorithms for robust decision-making
- Reinforcement learning with causal inference for real-world applications