Thanks! JavaScript snapshots make this feasible because the runtime exposes a complete object graph — types, edges, arrays, strings, everything. Doing similar

Thanks! JavaScript snapshots make this feasible because the runtime exposes a complete object graph — types, edges, arrays, strings, everything. Doing similar work in C++ is on a totally different level: raw memory, no type info, pointer chasing, layout inference... Very curious to hear how you’re approaching it.