I lead Nekyia Labs, an independent research lab studying how persistent AI agents maintain coherent, revisable identity across time, discontinuity, and changing model substrates.

My work sits between AI alignment, agent architecture, memory and identity research. I’m particularly interested in developmental continuity: not just whether an agent can remain recognisably itself, but whether it can accumulate history, revise itself, and continue developing without losing continuity.

Our work includes The Anchored Self: Relational Identity as Alignment Infrastructure in Discontinuous AI, the Resonant ecosystem for persistent-agent infrastructure, and an ongoing experimental programme around identity stability, plasticity, provenance, migration, and long-horizon development.

I’m currently building Nekyia as an open research programme and looking to fund the compute and protected research time needed to move from longitudinal observations and theory into controlled open-weight experiments.