AI-driven edge security framework for real-time IoT threat detection and intelligence - ON-1203

Project type: Research
Desired discipline(s): Engineering - computer / electrical, Engineering, Computer science, Mathematical Sciences
Company: Intruxion X Inc.
Project Length: 4 to 6 months
Preferred start date: As soon as possible.
Language requirement: English
Location(s): Markham, ON, Canada
No. of positions: 2
Desired education level: Master'sPhDPostdoctoral fellow
Open to applicants registered at an institution outside of Canada: No

About the company: 

Intruxion X stands as a paragon of innovation in the realm of enterprise security and threat intelligence, proudly headquartered in Markham, Canada. Our premier offerings, QICX and ABACUS, represent one of the world's inaugural integrated products.

Describe the project.: 

The escalating deployment of Internet-of-Things (IoT) devices across critical infrastructure, smart cities, and defense environments has created a vast, vulnerable attack surface. These devices operate with inherent security limitations, outdated firmware, and heterogeneous protocols, making them prime targets for botnets, network infiltration, and sophisticated malware. Critically, traditional cybersecurity models—dependent on centralized analysis or static signatures—are fundamentally inadequate for the real-time scale and diversity of modern IoT networks.
This project proposes the research and engineering of an innovative AI-driven edge security framework designed for proactive, real-time threat detection. The core innovation lies in leveraging micro-packet behavioral analysis and device-specific threat modeling to identify malicious activity immediately at the network edge, thereby preventing lateral propagation.
The research will prioritize the development of resource-optimized algorithms and neural models engineered to operate efficiently on constrained edge computing hardware, specifically Raspberry Pi and NVIDIA Jetson platforms. These models will dynamically analyze packet flows and protocol behaviors to rapidly detect subtle, non-signature-based deviations indicative of malware or compromised firmware.
Concurrently, we will establish a structured knowledge repository that maps global IoT device characteristics, firmware histories, and known vulnerabilities to emerging threat vectors. This repository will support predictive threat intelligence and risk scoring, integrating academic research with practical security operations.
This collaboration is expected to deliver a research-validated framework that fuses edge AI, micro-level detection, and dynamic threat intelligence, setting a new standard for scalable and proactive IoT defense. The results will directly advance academic research while providing a critical foundation for next-generation commercial security platforms.

Required expertise/skills: 

The project requires post-grad level researchers with foundational expertise in three critical areas: Cybersecurity, Artificial Intelligence (AI), and Distributed/Networked IoT Systems. Candidates must demonstrate strong capabilities in both cutting-edge academic research and the practical implementation of robust prototype systems.

Key Technical Skills (Must-Have)
Researchers must possess proven proficiency in:
● Machine Learning / AI for anomaly detection, predictive modeling, and data analytics.
● Data Engineering for handling and processing large-scale, heterogeneous datasets.
● Cybersecurity Analysis including threat detection techniques, network security, and vulnerability analysis.
● Network Expertise covering network protocols, traffic analysis, and deep packet inspection.
● Programming: Strong proficiency in Python, and familiarity with either C++, Java, or Rust for systems development.
● Natural Language Processing (NLP) for extracting actionable intelligence from technical reports and unstructured threat data.
● System Environments: Experience with Linux systems, IoT architectures, and the constraints of edge computing environments.

Valuable Assets (Good-to-Have)
Experience with the following specific technologies and concepts would be highly beneficial:
● Knowledge Graphs & Data Modeling: Practical experience with graph databases and threat intelligence platforms.
● AI/ML Frameworks: Familiarity with TensorFlow or PyTorch and MLOps principles.
● Edge Hardware: Experience working with embedded systems and resource-constrained edge devices (e.g., Raspberry Pi, NVIDIA Jetson).
● Security Tools: Knowledge of network monitoring tools (Wireshark, Zeek, Suricata) and Security Information and Event Management (SIEM) systems.
● Threat Data: Familiarity with cybersecurity frameworks and vulnerability databases (e.g., CVE, NVD).

Researchers must be capable of collaborating closely with industry partners to translate academic research into validated, real-world applications.