Evidence

How Thornveil proves it works

Every claim Thornveil makes is backed by working code, measured benchmarks, and where appropriate, cryptographically auditable reduction-to-practice. Five papers below. Public companion repositories on github.com/thornveil-ai for the rest.

Research papers

Peer-reviewable papers documenting the algorithms behind Thornveil's systems. HawkStack topology paper preparing arXiv submission; remaining four are technical reports.

Selective-Buffer Streaming Safety for AI Coding Agents

cs.CR 8 pages J. Morgan, 2026

Zeroth-Order Preference Optimization on 100B+ Quantized MoE Models via Live Inference API

cs.LG 7 pages J. Morgan, 2026

Compressed KV Cache Attention as a Plugin Backend

cs.DC 6 pages J. Morgan, 2026

Speculative Decoding Fails on Sparse MoE: A Negative Result and Practical Multi-Model Cascade Alternative

cs.SE 6 pages J. Morgan, 2026

RigRun: Complete Local AI Infrastructure on a Single GPU

cs.AI 5 pages J. Morgan, 2026

Trade-secret-grade IP across the portfolio

Each Thornveil system carries one or more proprietary subsystems where the engineering work materially differentiates it from open-source alternatives. Listed below: the named subsystem, what makes it defensible, and the cryptographic or empirical evidence that backs the claim.

Mycelium

Substitute-on-failure MoE expert dispatch

Day-27 RTP with HMAC-chained audit log, 2026-05-08

Auspex

Scope-as-code engagement compilation + 13-check Signet gate

Autonomous AD takeover validation with auditable replay package, 2026-05-16

RigRun

Compile-time classification-gated routing with cross-domain guard

Type-system enforced; 44 NIST 800-53 controls implemented

RigRun

8-signal confidence pipeline with adversarial self-audit

~4,220 LOC across 8 signals; cross-app parity in Go/Dart/TS

Signet

HMAC-chained tamper-evident audit ledger with RFC 3161 anchoring

2,400 LOC across chain.py + verifier.py + compactor.py; 7-state break taxonomy

HawkStack

Compute-aware neural topology recipe (3-parameter)

15 verified zoo checkpoints, R²=0.9895 power-law scaling fit on NUDT-SIRST

Pyros

24-layer adaptive control loop (PID + Holt + UCB1 + ε-greedy)

Race-clean CI across Linux/macOS/Windows; PAVA calibration validated vs sklearn

Meridian

Phase 10 trade-secret primitives — attest, secdef, beacon, vehicle-id, log

~5,500 LOC across 5 crates; not shipped in ArduPilot or PX4

Navigator

14-step agent-production pipeline with Opus calibration + 5-probe adversarial hardening

9 production domain-expert agents generated by the pipeline

10
Production systems
9
Trade-secret subsystems documented
5
Research papers