building things that compute
Independent researcher and engineer. I build engines, languages, and machine-learning systems, and I work on the mathematics of what survives perturbation.
The real structure of a system is the part that survives being perturbed. One test, run across fields that are not supposed to connect.
Perturbation robustness as the criterion for which structure in a system is real. Master theorems for the dominant invariant component, with a philosophical defence I call perturbative realism.
RH reframed as a positivity condition on a single observation operator over an admissible measurement space. The constraint is the statement; the construction is the open work. See predictions →
The naturals, integers, rationals, reals, and the primality indicator, all generated by exp(x) - log(y). Formalised in Lean 4 with no gaps left open.
The phantom sound modelled as a predictive loop running a fraction of a beat out of step, turning on a single coefficient. A clean model, and an honestly reported null in the first cohort.
Pitch and time as one continuous field rather than two categories, and real-time instruments built on it that reach transformations the separated tools cannot.
A single visibility-of-selector inequality for when any system – a model, a journal, a market – manufactures evidence for what it was already selecting toward.
~97 preprints with companion code · research.abhishek-shivakumar.com · predictions · interactive: stochastic resonance & UK heatwaves