ubugnu

Badreddine Bouzid ubugnu

Theoretical physicist and mathematician by training, full-stack developer by trade.

I did my PhD in theoretical physics at the Laboratoire de Physique Théorique d’Oran, Université d’Oran, on higher gauge theory: what 2-connections on 2-bundles look like when space is cut into a lattice. Today I build web platforms and self-hosted infrastructure at OrbitX.

Each point in the network is a field I work in. Lines join fields that feed into each other, and hatched triangles mark ideas where three of them meet.

Read field by field

Theoretical physics

Gauge theory is home ground: Yang–Mills fields, their quantization with BRST and Batalin–Vilkovisky methods, and Dirac’s analysis of constraints underneath. On the gravity side I worked with loop quantum gravity, spin foams, and the topological BF and Chern–Simons theories behind them.

My research [1] moves gauge theory one categorical level up. Put a 2-connection on a hypercubic lattice, with group elements on links and on plaquettes, and its transition laws come back in the continuum limit, such as

[1] B. Bouzid and M. Tahiri, 2-connections, a lattice point of view, arXiv:1803.00529 [hep-th].

    Mathematics

    Higher gauge theory runs on higher category theory, so I learned what lies beneath it: 2-categories, crossed modules, bundles, 2-bundles and gerbes. Around them sit the topology, geometry and algebra I use, from Lie theory and symplectic geometry to homological algebra and algebraic geometry.

      Logic and type theory

      Here mathematics turns into programming. Homotopy type theory reads types as spaces and equalities as paths, while the λ-calculus and its type systems are the theory behind the functional languages I write.

        Software engineering

        I build full-stack web and mobile apps and the infrastructure they run on. At OrbitX that means a sovereign workspace organizations host themselves: patched open-source apps, Ansible collections, rootless Podman, Caddy, Kubernetes, and Prometheus with Grafana.

          Machine learning

          I know the foundations of deep learning and how transformers and LLM agents work, and I follow the corner of the field closest to physics: physics-informed networks, and equivariant architectures that build symmetry in the way gauge theory does.

            Contact

            Write to me about physics, mathematics, or something you want built.

            The drawing borrows from my research. A 2-connection on a lattice carries data on links and on plaquettes; here fields sit on points, links join fields that feed each other, and hatched plaquettes mark where three meet. Set in Computer Modern, drawn with three.js.