AI-native risk & investment research
Kinetic AlphaGenerating excess return through the movement of money.
A working portfolio of risk and investment solutions built fast with AI. The throughline: capital that moves intelligently compounds, the scarce input is mental capital, and data ecosystems — increasingly on-chain — are the foundation that lets any of it work at all.
The dashboards
Two working tools, both shipped fast with AI.
ICE energy futures contract decomposition
Every ICE energy contract — crude, gas, power, refined, NGLs, coal, emissions — decomposed in your browser into dated factor legs. KPI strip, sector breakdown, scenario risk, multi-portfolio compare. A Python reference engine and a JavaScript browser engine that match cell-for-cell.
- Factors
- 121
- Contracts
- 107
- Py ↔ JS parity
- 32 / 32
Perps × predictive — cross-asset margin analytics
Bitcoin perp + Kalshi binary strips on the same underlying. SPX perp + Kalshi SPX binaries. Net delta, PnL curve, implied vol surface, live Kalshi feed. Plus an interactive 5,000-path Monte-Carlo of the 8-cluster portfolio margin framework, complete with the binding-constraint KPI that flips between 'Correlation' and 'Top-2 floor' as you drag the systematic-correlation sliders.
- MC paths
- 5,000
- Clusters
- 8
- Underlyings
- BTC + SPX
On deck: political-event cross-margining (elections, Clarity Act, war scenarios), ETH + SOL underlyings, full vol-surface fit, on-chain data layer for both dashboards.
Latest research
The writing behind the work.
Perps come onshore: what the CFTC's May 29 approvals change about contract design
The CFTC approved Kalshi's BTCPERP as the first US-regulated perpetual, issued a policy statement on listing perps, and cleared a Coinbase pathway to Deribit — all in the same 48-hour window. A walk through what a perp actually is, what knobs designers turn, and how the offshore (Hyperliquid HIP-3) and onshore (Kalshi DCM) paradigms compare.
The thesis
Four convictions that drive the work.
Excess return follows the movement of money
Capital that sits doesn't compound. Capital that moves intelligently — across venues, tenors, structures, and asset classes — does. Every project here is, at root, an attempt to identify a structural inefficiency in how capital is priced, margined, or routed, and turn it into a position you can defend.
Mental capital is the scarce input
Time and attention are the binding constraint on every real investment process. The dashboards here are built to push routine analysis down to seconds, so judgment is spent where judgment actually matters. Reducing the cognitive overhead of running a book is itself an alpha source.
You are the developer
AI has collapsed the time-to-solution curve. The projects here were each built in days, not quarters — by one person, end-to-end. That capacity is now in the hands of anyone who can describe what they want. The most important change in finance this decade isn't a new model; it's who gets to build them.
Data ecosystems are the foundation
Every financial system is downstream of its data layer. Open, verifiable, machine-readable data — increasingly mediated by blockchains — is what makes the next generation of risk and investment solutions buildable at all. Capturing value in modern finance starts with capturing the data ecosystem underneath it.
Building something AI-native in finance?
Margining, clearing, risk decomposition, on-chain data, agent-built research workflows — happy to compare notes or take a look at the problem.