What I know, and how it combines
I work across a handful of overlapping fields, and most of what I do lives where they meet. The chord below traces how these skills actually combine inside my papers and projects: every ribbon is two areas used together in the same piece of work. Switch to the full map for the complete inventory, from published depth to what I am still reading into. It is all written out below too, grouped by field, if you would rather skim.
Each ribbon links two skills used together in the same paper or project. Hover a skill or a project to isolate it; drag to pan, scroll to zoom. Colour marks the field, fill shows how deeply I know each skill. Click a field label to fold it; drag to pan, scroll to zoom.
Open interactive ↗ Interactive chord ↗ Full skill map ↗
The interactive maps open full-screen and work best on a wider display.
Continual & Adaptive Learning
Continual Learning, Online / Streaming Learning, Few-shot Learning, Open-World CL (OWCL), Novelty / Open-set Detection, Rehearsal-free Consolidation, Metaplasticity, Test-time & Domain Adaptation, Meta-learning, Replay / Rehearsal Methods
Embodied AI & Robot Learning
Robot Perception, Robot Learning, Vision-Language-Action (VLA), World Models, Predictive Representations, Closed-loop / Sensorimotor Control, Tactile / Multi-modal Fusion, Imitation / Behaviour Cloning, Reinforcement Learning, Diffusion Policies
Neuromorphic & HW Co-Design
Neuromorphic Computing, Intel Loihi 1 & 2, Spiking Neural Networks (SNN), Event-driven Computation, Near-memory / Dataflow HW, Fixed-point Arithmetic, Latency / Energy Profiling, Cross-platform Benchmarking, HW-aware Sparsity / Quantization, NVIDIA Jetson / TensorRT / ONNX, CUDA
Computer Vision & Perception
Computer Vision, Event Cameras, Object Recognition / Detection, Real-time Perception, Spatio-temporal Networks, Action Recognition, Multi-modal (Vision + Tactile), OpenCV
NeuroAI & Bio-Inspired Learning
Computational Neuroscience, Local & Greedy Learning Rules, Predictive Coding, Three-factor & Hebbian Rules, Neuro-inspired Control, Backprop Alternatives, Forward-Forward / Equilibrium Prop, Predictive Coding for Transformers
Deep Learning & ML Foundations
PyTorch, CNNs, Semi-supervised Learning, Self-supervised Learning (V-JEPA-style), Scikit-learn, Vision Transformers (ViT), Transformers (DINOv2, VLA backbones), RNN / GRU, Evolutionary / Genetic Algorithms, Foundation-Model / VLA Literacy
Programming, Systems & Tooling
Python, Git, Unix / Linux, MATLAB, ROS / ROS 2 / YARP, Gazebo / MuJoCo, Docker, C++ / Java
Application Domains & Interests
Real-time, low-power, on-device autonomy, Drone vision & on-board adaptive AI, Infrastructure inspection, Robotics (healthcare, household, industrial, agriculture), Edge AI / AR / VR, Physical AI & real-world systems, Frugal & distributed AI, Predictive maintenance