Sylvester Kaczmarek

Safe, secure AI and autonomy for critical systems, space, and robotics.

Technical Skills & Expertise

This section outlines my technical skills and expertise across AI and machine learning, cybersecurity, autonomous systems, space technologies, robotics, secure communications, software engineering, and critical systems.

Core Cybersecurity Domains

  • Network Security: Design, implementation, and management of secure network architectures; Firewall configuration and rule optimization (Cisco, Juniper, Palo Alto, pfSense); IDS/IPS deployment and tuning (Snort, Suricata); Network segmentation and micro-segmentation; VPN and secure remote access solutions (IPSec, OpenVPN, WireGuard), including considerations for high-latency environments.
  • System Security: Operating system hardening (Linux, Windows, RTOS); Secure configuration management; Vulnerability management; Endpoint security (EDR/XDR implementation); Identity and Access Management (IAM) principles.
  • Cloud Security: Cloud security principles (AWS, Azure, GCP) for ground segment infrastructure and data processing; Configuration of security groups and network ACLs; Identity management in the cloud; Understanding of container security (Docker, Kubernetes).
  • Application Security: Secure software development lifecycle (SSDLC) principles, including secure coding practices for resource-constrained systems; Understanding of common web vulnerabilities (OWASP Top 10); Static and Dynamic Application Security Testing (SAST/DAST) concepts.

Security Operations & Assessment

  • Vulnerability Assessment & Penetration Testing: Utilizing tools like Nessus, OpenVAS, Metasploit, Burp Suite for comprehensive security testing; Manual penetration testing techniques; Reporting and remediation guidance, including RF-specific testing for communication links.
  • Security Auditing & Monitoring: Log analysis and correlation; SIEM configuration and usage (ELK Stack, Splunk); Network traffic analysis (Wireshark, tcpdump, Zeek); Security audits based on frameworks like NIST and ISO 27001.
  • Incident Response: Incident handling lifecycle (preparation, detection, analysis, containment, eradication, recovery, post-incident); Coordination of response efforts, including scenarios specific to space asset compromise or interference; Digital forensics fundamentals.

Cryptography & Secure Communications

  • Encryption Technologies: Implementation and management of symmetric/asymmetric encryption (AES, RSA), hashing algorithms (SHA-2/3), digital signatures, and PKI, including lightweight cryptography suitable for space systems.
  • Secure Protocols: Deep understanding and implementation of secure communication protocols (TLS 1.3, DTLS, SSH, IPSec, QUIC); Protocol analysis and design for specialized environments (e.g., CCSDS Space Data Link Security (SDLS) protocols, delay-tolerant networking (DTN) security).
  • Data Integrity: Ensuring data integrity through techniques like HMACs, authenticated encryption (AES-GCM), and error correction/detection codes relevant to noisy space channels.
  • Key Management: Best practices for cryptographic key generation, distribution, storage, and lifecycle management, adapted for distributed and remote space assets.

AI, Machine Learning & Autonomous Systems

  • Machine Learning & Deep Learning: Design, training, evaluation, and deployment of supervised and unsupervised learning systems using PyTorch, TensorFlow, scikit-learn, NumPy, SciPy, and pandas, including anomaly detection and classification for safety-critical and resource-constrained environments.
  • Safe, Secure & Trustworthy AI: AI safety and alignment, adversarial robustness, runtime assurance, failure detection, calibrated decision support, and evaluation of systems operating under uncertainty, adaptation, and adversarial pressure.
  • Large Language Models: Secure LLM design and evaluation, model and application-layer security, robustness considerations, responsible deployment, and assurance for high-risk or sensitive use cases.
  • Interpretability & Explainability: Model inspection and explanation techniques including SHAP, Integrated Gradients, Layer-wise Relevance Propagation, feature attribution, and human-in-the-loop assurance.
  • Adaptive & Continual Learning: Continual learning, adaptive thresholds, metaplasticity, alignment persistence, resistance to catastrophic forgetting, and controlled adaptation under sequential updates.
  • Neuromorphic Computing: Spiking neural networks, temporal coding, adaptive neuromorphic learning, robustness and anomaly detection, with emphasis on low-power edge inference and autonomous systems.
  • Edge & Onboard AI: Model compression, quantization, ONNX-based workflows, CUDA/TensorRT acceleration, and deployment of AI models to constrained edge and onboard computing environments.

Specialized Security Expertise

  • Space & Satellite Systems Security:
    • Satellite Bus & Payload Security: Securing onboard computers, operating systems (RTOS, embedded Linux), and payload instruments.
    • TT&C Security: Protecting Telemetry, Tracking, and Command links against unauthorized access, spoofing, and jamming (including RF layer security analysis).
    • Ground Segment Security: Securing mission operations centers (MOCs), ground stations, communication networks, data processing pipelines, and cloud infrastructure supporting space missions.
    • Launch Segment Security: Awareness of security considerations during integration, testing, and launch phases.
    • Space Data System Standards: Deep familiarity with CCSDS standards (e.g., TC/TM, AOS, SDLS) and their security implications and extensions.
    • Space Supply Chain Security: Understanding risks and mitigation strategies for hardware and software components used in space systems.
    • Resilience & Fault Tolerance: Designing security architectures that accommodate the harsh space environment (radiation effects) and operational needs for resilience.
    • Regulatory Awareness: Familiarity with space cybersecurity guidelines and directives (e.g., NIST SP 800-235, Space Policy Directives).
  • AI/ML Security:
    • Threat Modeling for AI Systems: Identifying unique vulnerabilities and attack vectors in AI/ML pipelines.
    • Adversarial Machine Learning: Understanding and defending against evasion, poisoning, and inference attacks.
    • Secure AI Development & Operations (MLSecOps): Securing ML pipelines, training data, model storage, and deployment environments.
    • AI Governance & Responsible AI: Implementing security controls aligned with ethical AI principles, fairness, transparency, and accountability.
    • Privacy-Preserving ML: Familiarity with techniques like federated learning, differential privacy, and homomorphic encryption in ML contexts.
    • AI Red Teaming: Assessing the security posture of AI systems through simulated attacks.
  • Robotics & UAV Security: Securing control systems and communication links for robotic and unmanned systems; Protection against hijacking and sensor spoofing; Embedded system security for robotic platforms.
  • Embedded & Control Systems Security: Real-Time Operating System (RTOS) security considerations; Securing embedded Linux; Hardware security module (HSM) concepts; OT/ICS security fundamentals.
  • Geospatial Intelligence Security: Applying security principles to GIS data handling and analysis workflows (ArcGIS); Secure transmission and storage of geospatial data.

Autonomous, Robotic & Space Systems

  • Autonomous Systems: Architecture and evaluation of autonomous decision-making systems, including monitoring, verification, human oversight, fault handling, and operation where continuous supervision cannot be assumed.
  • Robotics: AI/ML for robotic and physical systems, ROS 2, perception and decision pipelines, safety monitoring, and integration of autonomous software with real-world platforms.
  • Space Systems & Onboard Autonomy: Spacecraft onboard intelligence, payload autonomy, mission operations, telemetry and command systems, edge processing, and autonomous operation in remote and communication-constrained environments.
  • Hardware-in-the-Loop & Validation: Hardware-in-the-loop evaluation, simulation, test automation, verification and validation, and reproducible experimental workflows for autonomous and embedded systems.
  • Spaceflight Analysis & Simulation: Orbital and mission simulation, autonomous spacecraft operations, and analysis of mission-relevant dynamics and operational constraints.

Programming & Automation

  • Python: Security scripting for automation, analysis, tool development, AI/ML security tasks, and interacting with APIs (including space systems APIs where applicable).
  • C++: Performance-oriented software development for systems, simulation, robotics, embedded computing, and computationally constrained applications.
  • Bash: Linux/Unix shell scripting for system administration, automation, and security tasks.
  • Rust: Developing performance-critical and memory-safe security tools and applications, suitable for embedded and space system components.
  • Scientific & ML Computing: PyTorch, TensorFlow, scikit-learn, NumPy, SciPy, pandas, OpenCV, Jupyter, and reproducible research workflows.
  • Deployment & Engineering Tooling: Linux, Docker, Git/GitHub, GitHub Actions, ONNX, CUDA, TensorRT, CI/CD workflows, and LaTeX-based technical documentation.

Governance, Risk & Compliance (GRC)

  • Security Frameworks: Implementation, assessment, and alignment with ISO 27001 and the NIST Cybersecurity Framework (CSF), including AI-specific risk considerations (e.g., NIST AI RMF) and space-specific adaptations.
  • Regulations: Familiarity with key data protection and compliance requirements (GDPR, HIPAA, SOC 2 principles), emerging AI regulations, and relevant space directives/guidelines.
  • Risk Management: Risk assessment methodologies; Threat modeling (including specific AI threat models like MITRE ATLAS and space-specific threat vectors).

Feel free to reach out via LinkedIn to discuss collaborations or professional inquiries.