Engineering accounts, economic research, founder notes.
What we build, what it costs to run, and how the mechanisms underneath actually work. Written by the founders. Claims anchored to repositories, DOIs, and mainnet deployments.
We shipped an order-book DEX. AMMs won anyway. Here's the math.
Our matching engine worked; the maker economics didn't. The doom-loop math of thin-venue market making, and why order books retreated to perps — where they now dominate.
No threat model, no privacy work
Privacy tech defends against someone specific. Without a named adversary, 'private' is a mood — and we decline the build. What the one-page threat model contains.
What institutional stablecoin engineering involves after GENIUS and MiCA
The regulation moved stablecoins from token contracts to systems engineering. The GENIUS/MiCA requirements table, and the four workstreams they actually mandate.
The usefulness gap: what proof-of-useful-work actually pays for
The test of useful work: would anyone buy it at the price the network pays? Bittensor's post-halving emissions vs open-market GPU prices, and the gap nobody publishes.
The price of not trusting: a working cost model for ZK light clients
One formula, public inputs, a lopsided answer: verification gas outweighs proving 25-100x on L1. Where the money goes and when proofs are worth buying.
Mainnet is the halfway point: the real cost structure of keeping a chain alive
Launch budgets buy a mainnet; nobody budgets the decade after. Kadena's shutdown and Protocol Guild's funding data put numbers on the five costs that survive genesis.
Anatomy of a bond-backed stablecoin: how Evidentia works
A fiat-backed stablecoin is an accounting system; a bond-backed one is a term-structure machine. Per-bond ERC-1155 collateral, the mint formula, and where the yield really comes from.
One proof for any leaf: OR-aggregation Merkle inclusion proofs
The engineering translation of our peer-reviewed result: aggregate inclusion proofs with OR logic and proof size stops depending on the tree.
Local-currency stablecoins are an emerging-market product, not a dollar clone
Who actually needs hryvnia, lira, and dong onchain, what the World Bank's corridor data says, and why currency-matched bond backing is the design that works.
A ZK light client is a consensus reimplementation
The engineering account of the NEAR→Ethereum client built under a Wormhole Foundation contributor grant: five-block finality, ed25519 in-circuit, recursion strategy.
Efficient and Universal Merkle Tree Inclusion Proofs via OR Aggregation
Batch inclusion proofs whose size is independent of leaf count, via OR-logic aggregation.
Enhanced Security and Efficiency in Blockchain With Aggregated Zero-Knowledge Proof Mechanisms
Aggregating zero-knowledge proofs to cut verification overhead in blockchain systems.
Merkle Trees in Blockchain: A Study of Collision Probability and Security Implications
Quantifying collision probability in Merkle constructions and what it implies for security margins.
Adaptive Merkle Trees for Enhanced Blockchain Scalability
Restructuring Merkle trees adaptively as chain state grows.
Solving Blockchain Scalability Problem Using ZK-SNARK
Compressing blockchain state verification with ZK-SNARKs.