AI Builds Custom Decision-Making Software in About an Hour Instead of Months
NTU Singapore’s LACE framework turns a plain-language description of a scheduling or routing problem into a tested portfolio of optimisation algorithms in about an hour. We explain its two stages, the benchmark results against FunSearch, EoH and Gurobi, and use the team’s published figure data to check the hour and $10 claims across four AI models, then set out who could use it and its limits.
