The same discovery engine, rendering orbital structure · KODON
The same discovery engine — deterministic, exact-or-abstain, held-out validated — pointed at the sky, the spacecraft exploring it, tomorrow's weather, and the machines that can't afford to fail quietly. Every number on this page traces to a run you can reproduce.
The same discovery engine, rendering orbital structure · KODON
Same rule everywhere: prove it exactly, or say nothing. No exceptions for difficulty.
Nine of the deepest open problems in physics and math. One deterministic engine, no LLM anywhere in the loop. Every verdict below is verbatim — real survey data, real telescope catalogues, the entire OEIS — and wrong zero times across ~28 rounds.
| Mystery | What NOESIS did | Verdict |
|---|---|---|
| Particle masses (Koide) | Compressed charged-lepton masses to one scale + two exact constants | r=√2 @ 0.42σ, p≈0.0036 — survives |
| Neutrino masses | Exact-Koide + NuFIT 6.0 splittings → absolute masses | Σmν = 0.0592 ± 0.0002 eV, DESI DR3 decides |
| Dark energy | Full-covariance audit of DESI BAO (13 meas.) + Pantheon+ SNe (1,580) | 3–4σ claim lives only in the combination |
| Hubble tension | Solved the full SH0ES ladder from raw matrices (3,492 obs) | H0 = 73.043 ± 1.007, matches paper exactly |
| Dark matter | RAR over SPARC (2,696 pts, reproduces McGaugh exactly) | a0 ≈ cH0/2π real, undecidable at current systematics |
| Prime structure | Spectral audit of 100,000 real Riemann zeros | Montgomery–Odlyzko replicated: GUE beats Poisson 14.9× |
| Exoplanet law | 1,229 planets, 877 adjacent pairs, MC-calibrated | Resonance piling real (11.6% vs 4.6% null, p<0.0005) |
| GW background | Real NANOGrav 15yr chains (502,300 samples) | γ = 3.34, SMBH prediction ~3σ tension |
The deriver ladder swept all 397,116 OEIS sequences: 24,604 derived with exact held-out verification, distilled to 21 substantive uncatalogued recurrences. The same funnel executed its own plausible-but-wrong candidates without mercy — TREE(i) thresholds secretly constant, palindromic counts secretly zero, a 1992 conjectured generating function hiding in a Maple comment.
DESI's dark-energy signal swings with which supernova set you add — 2.8σ with Pantheon+, 3.8σ with Union3, 4.2σ with DES-Y5. That's exactly where an independent, calibrated audit earns its keep: not another posterior, but an instrument that regression-tests each release.
A published, zero-dependency Rust crate that watches a time series' own statistical structure and flags the moment it starts to break down — no training, no black box. Same math, no retuning, across spacecraft, bearings, cardiac signals, and DNA.
no_std.struktura generate --cfs/--fprime/--ros emits a complete flight-software monitor.An independent contributor's engagement with NASA's public open-source flight-software repos — reported exactly as merged or open, nothing rounded up.
| Repo | What | Status |
|---|---|---|
| nasa/fprime | PR #5556 · FileDownlink buffer fix | merged |
| nasa/fprime | PRs #5760–#5771 · bug fixes, defense-in-depth | open, in review |
| nasa/ogma | PR #552 · mergeSpecs data-loss fix | open, awaiting review |
| nasa/ogma | Discussion #557 · DFA custom-codegen template | open |
| nasa/cFS | Proposal #1096 · Structural Health app | open, NASA feedback received |
Every number on this page traces to a reproducible run on public data. For DESI: a standing regression audit on future releases, zero integration burden. For Struktura: run it on your own telemetry.