AI Engineer & Researcher
Machine Learning · AI Systems · Computer Vision · NLP · Privacy-Preserving AI
|
End-to-end pipelines, model development, evaluation, deployment-oriented workflows, and reproducible experiments. |
CNN-based image classification, object detection, medical imaging workflows, and Persian license-plate recognition. |
|
Text preprocessing, sentiment analysis, classification, and practical workflows around modern language-model tooling. |
Federated learning, differential privacy, and ML system design under privacy constraints. |
|
Memory behavior, benchmarking, systems measurement, and understanding software/hardware interaction. |
Data cleaning, feature analysis, visualization, business-oriented exploration, and statistical reasoning. |
|
End-to-end computer-vision workflows for Persian license-plate detection and recognition, including YOLO-based experiments. |
A hybrid intelligent-control project combining time-series forecasting with fuzzy-logic reasoning for greenhouse environments. |
|
Persian food-delivery review sentiment analysis with preprocessing, machine learning, and a deployable application workflow. |
Regression pipeline covering preprocessing, model comparison, tuning, evaluation, learning curves, and Docker-based execution. |
|
Classification workflow for heart-disease risk analysis using multiple machine-learning algorithms, notebooks, and Docker support. |
Linux/C++17 tooling for memory latency, cache hierarchy behavior, memory bandwidth, access patterns, and system performance. |
Selected work is intentionally curated; the full project collection is available in the repositories below.
| GNNs Graph learning |
LLMs Language models |
FL Federated learning |
DP Differential privacy |
MLOps Reliable pipelines |
Mehrab Jalilian · مهراب جلیلیان
AI Engineer · Researcher · Computer Engineering
