Ambitious continuation of Hilbert's 16th problem, part (a), degree 8, on https://github.com/wustep/maths. Folder problems/hilbert16-degree-8/ already has a 2,384 lower bound (seventeen T-curves outside the 2,367 of arXiv:2602.06888v3). Fetch that paper before claiming anything. Replay the existing census/verifiers first. Do not regress 2,384. Work in the next compute/qN/. Hunt a large jump: more real schemes than 2,384, a finished deep-nest decision, or a completed sign-vector-stratum sweep. Isolated timeouts are residue. Open a PR.