About
Rosetta Alpha
Institutional reasoning layer
Built on the
All Weather discipline.
Rosetta Alpha is fundamentally inspired by Ray Dalio's pioneering work on the All Weather strategy and Bridgewater's culture of radical truth. By combining multi-agent LLM reasoning with decentralized prediction markets, we've reimagined Dalio's principles for the AI era.
Most AI investment platforms speak one language and see one market. Rosetta Alpha runs five regional analysts, each native to their market's culture, language, and data sources, then settles their conviction on-chain as verifiable prediction markets.
Balanced Economic Exposure
Commodities
Cash
CommoditiesRising Inflation
Nominal BondsFalling Inflation
Ray Dalio's strategy succeeds because it acknowledges that no single asset class captures every economic regime. Rosetta Alpha extends that discipline to the AI layer: maintaining structural equilibrium across the four distinct economic environments that drive asset returns.
Rising Growth
Equities & Commodities
Positive growth surprises favor ownership of the productive economy.
Falling Growth
Nominal Bonds & Cash
Economic contraction increases the value of safe, fixed-return assets.
Rising Inflation
ILBs & Commodities
Currency devaluation favors inflation-linked bonds and tangible assets.
Falling Inflation
Equities & Nominal Bonds
Deflationary environments increase real returns on financial capital.
The Pipeline
- 01
Regional ingest
Each desk pulls market data, filings, and news in the local language.
- 02
Multi-agent reasoning
Specialized analysts (fundamental, technical, sentiment, macro) produce structured ReasoningBlocks.
- 03
Thesis synthesis
A Portfolio Manager agent reconciles agent outputs into a single direction + conviction.
- 04
Provenance
The full thesis is hashed, pinned to IPFS, and the hash is recorded on Arc L1.
- 05
Market settlement
A binary prediction market is opened on the thesis question. Settlement is autonomous.