Machine Learning Engineer

MSc, Artificial Intelligence and Robotics · Amirkabir University of Technology, Tehran

Thesis: "Designing an Embedded System for Activity Recognition Using a Surveillance Camera"

Real-Time Activity & Intention Recognition

Built an end-to-end video-to-decision pipeline: compresses a motion clip into a single time-encoded image, classified by a lightweight AlexNet-style CNN. Designed for embedded use: the CNN and image pipeline were simulated with Xilinx simulators for a ZCU102 FPGA board.

Created the training dataset from scratch (no public dataset existed).

First author of the arXiv preprint.

Accuracy in the paper: 98.78% on my dataset, 97.95% on KTH, 100% on Weizmann, 78.48% on HMDB-51. · reported in the 2021 preprint

• arXiv preprint
• Medium write-up

AI/ML Project Reviewer · Abadan University of Medical Science, Abadan

Reviewed deep-learning projects in medical image analysis (model architecture, training procedure, evaluation metrics).

Deep Learning Mentor and Teaching Assistant (Part-Time) · Maktabkhooneh, Tehran

Taught and mentored 100+ students in Reinforcement Learning, CNNs, and Deep Learning.

Co-developed the course curriculum with industry examples.

so I spent 4.5 years as a Java engineer in Munich learning that side properly: TDD, CI/CD, code quality gates, Docker.

• Mar 2022 - Mar 2025Software Developer · TESTIFI GmbH, Munich: Built core QualityOps features with TDD from stakeholder requirements.
• Mar 2025 - PresentFull Stack Developer · Convien GmbH, Munich: Lead code-quality initiatives against code smells, vulnerabilities, and technical debt.