Optimization work, most of it open source. The demos compute their results live — nothing on this site is a recording.
Every result here is computed when you ask for it. Each page says where it runs.
Robust optimization
A sampling-based method that hands you a Pareto frontier of robust solutions and lets you choose the trade-off after seeing the options, instead of committing to an uncertainty budget before you know what it costs.
Bandwidth Packing cases matched or beaten: 92.5% (as published, ESWA 2022)
Robust optimization
A continuous, normalized robustness measure for optimization under interval uncertainty — one number in [0,1] that ranks feasible solutions by how well they resist the uncertainty around them, with an exact closed form and a Monte-Carlo estimator that agree.
Measure range: η ∈ [0, 1], continuous
Metaheuristics
A public repository of mine contains 201 committed simulated-annealing runs, every one of them infeasible. This takes the failure apart, isolates its three causes, and reaches a few percent of the proven optimum in milliseconds.
Committed runs: 201
Mixed-integer programming
Fit the most people into a room subject to a minimum-separation rule — maximum independent set on a conflict graph, solved to proven optimality in the browser, on a floor plan you can rearrange.
Capacity at 1 m separation: 36 of 36 seats
Work worth reading about, without a live version.
Constraint programming
A verified database of Costas arrays for orders 2–100, algebraic generators built over finite fields, and CP/SAT/LP experiments against the smallest orders where nobody has ever found one — or proved none exists.
Arrays verified on every CI run: 9,217