I’m a computer engineering student working across machine learning, graph learning, privacy-preserving AI, language models, and practical ML systems.
I’m interested in the layer between research and deployment: building models, designing reliable pipelines, and turning technical ideas into systems that can actually be used.
Research should survive contact with reality.
That means reproducible experiments, measurable results, clean interfaces, and engineering decisions that hold up beyond the first demo.Graph learning for network intrusion detection and temporal systems.
Differential privacy and aggregation under distributed constraints.
Retrieval, agent workflows and grounded generation.
Deployment-minded pipelines, APIs and evaluation loops.
Linux/C++17 tooling for measuring memory latency, cache hierarchy behavior, and memory bandwidth.
VIEW REPOSITORY ↗An end-to-end computer-vision workflow for Persian license plate detection, character segmentation, and recognition.
VIEW REPOSITORY ↗Persian sentiment classification with Hazm preprocessing, classical ML, Transformers, FastAPI, and Docker.
VIEW REPOSITORY ↗Image-classification experiments covering preprocessing, augmentation, evaluation, and error analysis.
VIEW REPOSITORY ↗Notebook-based machine-learning work centered on a brain-disease classification workflow.
VIEW REPOSITORY ↗Sales-data analysis covering cleaning, cancellation patterns, customer analysis, time-based trends, and visualization.
VIEW REPOSITORY ↗Federated learning and differential privacy introduce constraints around data movement, aggregation, evaluation, and reproducibility. The interesting engineering begins where those constraints meet the pipeline.
A model that works in a notebook is only the beginning. Reproducible preprocessing, stable inference, validation, packaging, and observability are what turn an experiment into a service.
Memory behavior is shaped by cache levels, access patterns, bandwidth, and the distance between the CPU and data. A useful benchmark makes those effects measurable without pretending one number explains the entire system.
I'm always open to discussing new projects, creative ideas or opportunities to be part of your visions.
AI / ML Systems Web Applications
Research & Prototyping Technical Consulting
Let's create something impactful together.
