Software / Security engineer working at the intersection of reverse engineering, machine learning, and algorithms on hard, structured data — binaries, graphs, network traffic, recommendation system — and increasingly on medical imaging.
M.Sc. in Computer Science (Software Engineering), Ferdowsi University of Mashhad. Three peer-reviewed publications; 10+ years building production systems in automotive security and applied ML.
- 🔬 Research interests: AI for cybersecurity (LLMs & RAG for automated binary analysis), binary reverse engineering of automotive ECUs, anomaly detection & concept drift, machine learning on structured/graph data, high-performance & parallel computing.
- 🌐 Portfolio & CV: https://emad-mahmodi.github.io
- 🎓 Google Scholar: (Scholar profile link)
- 📫 Contact: emad.mahmodi.eng@gmail.com
-
A drift-aware adaptive method based on minimum uncertainty for anomaly detection in social networking. E. Mahmodi, H. Sadoghi Yazdi, A. Ghaemi Bafghi. Expert Systems with Applications, 2020. doi:10.1016/j.eswa.2020.113881 · code → drift-aware anomaly detection
-
UDIS: Enhancing collaborative filtering with fusion of dimensionality reduction and semantic similarity. H. Koohi, Z. Kobti, T. Farzi, E. Mahmodi. Electronics (MDPI), 2024. doi:10.3390/electronics13204073
-
A novel trust computation method based on user ratings to improve the recommendation. R. Barzegar Nozari, H. Koohi, E. Mahmodi. International Journal of Engineering, 2020. doi:10.5829/ije.2020.33.03c.02
| Project | What it is |
|---|---|
| drift-aware anomaly detection | Uncertainty-weighted ensemble for anomaly detection in imbalanced, drifting data streams — reference code for the ESWA 2020 paper. |
| phishing detection (URL obfuscation) | Rule-based phishing detection robust to typosquatting, with dataset tooling and reimplemented baselines. M.Sc. thesis. |
Much of my work is proprietary and cannot be open-sourced. A representative selection — described at the capability level, code and product details omitted:
Automotive security & reverse engineering
- Built a high-fidelity PowerPC (PPC) emulator for full ECU firmware analysis and Seed/Key security research.
- Reverse-engineered multi-core ECU firmware (binary de-obfuscation, logic reconstruction, CAN-bus monitoring, security-access analysis) across several ECU families.
- Developed a Qt/QML + Python protocol analyzer for UDS and OBD-II diagnostics over logic-analyzer captures.
- Security auditing of automotive diagnostic tools and Android/Bluetooth OBD2 systems (anti-debugging, V-table obfuscation, runtime memory protection).
AI for binary analysis
- RAG pipeline for LLM-assisted binary understanding — assembly-to-C translation, function-name recovery, type/struct inference, decompilation support.
Detailed write-ups (non-confidential) are on my portfolio.
Languages · C / C++ · Python · Java · MATLAB · Assembly (PPC/x86) ML / Data · PyTorch · scikit-learn · anomaly detection · concept drift · ensemble methods · RAG / LLM pipelines Systems / Security · reverse engineering · symbolic execution (SymCC/SymQEMU) · binary analysis · CAN / UDS / OBD-II · Qt/QML Parallel / HPC · CUDA · OpenMP (in progress)
⭐ Most of my strongest work lives inside industry projects under NDA — the repositories here are the publishable slice. Happy to discuss the rest directly.