Post Fiat Research

Blog

Post Fiat research notes, governance experiments, and technical updates.

A Proposal for Better Private FX Settlement

Foreign exchange settles on rails that leak size and carry Herstatt risk on every uncovered leg. We rebuild the argument from theory (settlement finality, Kyle’s model of information and impact, frequent batch auctions, dark-pool adverse selection) and propose a shielded frequent batch auction with atomic PvP over reserve-backed assets. New in this revision: a square-root law for internal netting, an information bound on what the externalized residual reveals, an equal-cost rule for batch length, and an unsentimental account of the two conditions the design still has to earn: a matcher nobody can see into, and a crowd.

Post Fiat Latency Series II: FastPay — Removing Consensus from Owned-Value Settlement

Series II implements a FastPay-style owned-value lane for Post Fiat: simple payments finalize on validator certificates, while consensus checkpoints them asynchronously. The result is measured on a controlled EU validator fleet and a six-validator cross-continent testnet — reaching 183 ms finality across three continents with post-quantum signatures — and is supported by a formal safety invariant, implementation notes, and an adversarial gate.

Proof of Disclosed Leverage

A zero-knowledge proof of a disclosed leverage profile: six legs, six verification domains, no hidden address disclosure. The proof commits spot, cash, gross perpetual notional, and Aave debt, with each line labeled by quantity and valuation evidence on a three-rung trust scale. It does not prove global solvency or completeness; it proves the disclosed account set reconciles exactly to the public buckets.

Pricing Counterparty Risk into the NAVCoin

A NAVCoin verifies what its venues report. Option markets price whether those venues survive. Multiply the two and every NAVCoin carries a live credit spread on-chain — turning the silent source of crypto yield into a number anyone can read.

The NAVCoin Proposal

A NAVCoin is a token whose unit value tracks the machine-verified net asset value of a reserve portfolio — not a peg. The settlement rail already runs on the Post Fiat devnet; this post proposes the verification layer, the privacy model, and the cross-chain mechanics, and is published before implementation so the design can be attacked first.

Post Fiat Latency Series I: Certified Finality vs. Validated Ledgers

A local one-host, six-validator study comparing Post Fiat L1 v2 certified applied-batch receipts with private rippled validated-ledger inclusion: 88.083 ms p50 for the tested Post Fiat full-vote path, plus an XRPL timer-compression matrix showing median gains and tail damage.

Cobalt on the Devnet: Implementing the Road Not Taken

Post Fiat has implemented Cobalt — the 2018 trust-evolution protocol Ripple proposed and never shipped — on its devnet, with a public, replayable evidence bundle. Adoption is an open governance question. Here is the problem Cobalt solves, what we built, how to verify it, and what it would mean under a control-based regulatory test.

Orchard/Halo2 Devnet Research Note

Orchard-style shielded accounting, the June 2026 Zcash Orchard remediation, and the corresponding Post Fiat devnet verifier boundary.

Viability of SGLang Replay: Cross-Hardware

An evidence note on SGLang replay across H200 and H100 NVL profiles: exact held-out packet replay, explicit prompt-drift quarantine, MLX capability control, machine receipts, comparison hashes, and shutdown receipts.

LLM Governance Replay: Cheap Public Verification Before Private Committees

A replay experiment on XRPL amendment history: deterministic LLM replay can make governance review cheaper by making first-pass claims public, typed, hash-bound, and contestable—not by replacing validator judgment.

Orchard Privacy Research

A Post Fiat privacy research note covering Orchard/Halo2 shielded payments, Railgun-style assurance, DID-bound providers, and why this material no longer belongs inside the validator-list whitepaper.