"I cannot guarantee a win — but I can engineer systems that make failure mathematically expensive, bounded, and observable."
| Role | Chief Engineer · Graceman |
| Academic | MSc Candidate, Research in Computer & Systems Engineering · Technische Universität Ilmenau |
| Research Outreach | davidgracemann@graceman.de |
| Engineering Outreach | gracemann365@gmail.com |
Graceman is a research engineering organisation built around one problem: constructing AI systems whose failure modes are formally bounded — not probabilistically hoped away.
The operational domain is autonomous systems under adversarial conditions. Reliable autonomy in denied, degraded, and operationally constrained environments is an open engineering problem. That is the north star.
| Codename | Cluster | Core Subjects |
|---|---|---|
[AML] |
Adversarial ML & Robustness | Adversarial attack and defence, OOD detection, reliability under sensor noise and degradation |
[RAS] |
Robotics & Autonomous Systems | ROS2, SLAM, path planning (A*, RRT, D*), sensor fusion, localisation |
[CV] |
Computer Vision | Object detection (YOLO-family), multi-object tracking, semantic segmentation, scene understanding |
[EAI] |
Edge AI & Embedded Inference | TensorRT, ONNX Runtime, OpenVINO, NVIDIA Jetson, INT8/FP16 quantisation |
[SE] |
Systems Engineering | Linux internals, distributed edge compute, node orchestration, tactical network stack |
[SPW] |
Signal Processing & Electronic Warfare | DSP fundamentals, radar signal processing, EW principles, FFT, adaptive filtering |
[HWE] |
Hardware & Embedded Systems | FPGA fundamentals, microcontroller programming, RTOS (FreeRTOS), PCIe interfaces |
[SRC] |
Secure & Resilient Communications | Mesh networking, tactical comms protocols, applied cryptography, secure enclaves |
[HPC] |
HPC & Low-Latency Inference | CUDA programming, kernel optimisation, latency profiling, memory bandwidth management |
| Codename | Cluster | Core Subjects |
|---|---|---|
[FMV] |
Formal Methods & Verification | TLA+, model checking, temporal logic, programme verification |
[CTR] |
Control Theory | Classical control, modern state-space (LQR, MPC), nonlinear systems |
[SYT] |
Systems Theory | Dynamical systems, Lyapunov stability, observability and controllability |
[OPT] |
Optimisation Theory | Convex optimisation, integer programming, Lagrangian methods |
[PSP] |
Probability & Stochastic Processes | Measure-theoretic probability, Markov chains, SDEs, martingales |
[LAD] |
Linear Algebra at Research Depth | Spectral theory, tensor algebra, matrix decompositions, operator theory |
[NUM] |
Numerical Methods | ODE/PDE solvers, numerical linear algebra, stability and convergence analysis |
[TCS] |
Theoretical Computer Science | Complexity theory, computability, algorithm analysis and lower bounds |
┌─────────────────────────────────────────┐
│ Mathematical Foundation │
│ Formal Methods · Control · Stochastic │
│ Optimisation · Linear Algebra │
└──────────────────┬──────────────────────┘
│
▼
┌─────────────────────────────────────────┐
│ Autonomous Systems & Defense AI │
│ │
│ Adversarial ML · Robotics & ROS2 │
│ Computer Vision · Edge AI & FPGA │
│ Signal Processing · Secure Comms │
└──────────────────┬──────────────────────┘
│
▼
┌─────────────────────────────────────────┐
│ GRACEMAN │
│ Foundational AI Research Lab │
│ │
│ AI systems whose failure modes are │
│ formally bounded, observable, and │
│ mathematically expensive to trigger. │
└─────────────────────────────────────────┘
Last Update · June 2026