Author: Cecilia (Qi) Feng, Founder of Their Foundry (4 June 2026)
The world has changed, and we have evolved with it. In the aftermath of the $19B flash crash, Their Foundry became a strategic tooling company, building an AI-native interpretive system for Macro-to-Onchain Liquidity Stress that delivers decision support at scale. Our mission is to help market structure stewards interpret macro signals early, position against onchain liquidity shocks ahead of time, and protect product and infrastructure continuity under stress.
It took time to arrive here. The path was not linear, but in retrospect it was increasingly inevitable. The market has already felt the force of macro shocks it cannot control. But while the pain is increasingly familiar, the tooling required to anticipate and position against downstream liquidity stress remains underdeveloped. One of the earliest signals of market pull came when market structure stewards began reasoning through the flash crash using our interpretive framework. As a result, a key missing category revealed itself to us: a systematized interpretive layer for macro-to-onchain liquidity stress.
This is our story: a journey rooted in the conviction that systems carry worldviews and shape possibilities, and in an interpretive lens that sees crypto as part of a larger interconnected system influenced by broader macro conditions. At the core of our work is a straightforward question: how do macro, policy, and geopolitical shocks propagate through onchain systems and become downstream liquidity stress — and how can decision-makers position early enough to reduce disruption?
Our Starting Point and a Distinctive Interpretive Lens on October 10th Flash Crash
Their Foundry began with a conviction that systems encode values, assumptions, and a worldview about how the world is organized — how infrastructure becomes culture, and how system design conditions the range of human and institutional possibilities available within it. That conviction shaped how we approached crypto from the beginning: as a live system of coordination, incentives, and market structure under changing conditions. Such orientation first led us into a thesis- and research-led company grounded in interdisciplinary research and crypto-native expertise in infrastructure and DeFi.
The $19 billion crypto flash crash on October 10, 2025 directly shaped our current direction. As we monitored the increasingly adverse macro environment leading into the event, its severe pro-cyclical liquidations made our conviction more concrete. The crash showed us that onchain financial systems do not simply reflect the conditions around them; their structure shapes how stress moves through them and how large the downstream consequences become. That, in turn, raised an existential question for us: when macro conditions turn hostile, how do onchain financial systems transmit shock, absorb it, or expel it?
In addition to the blast radius of the event, what drew us further into the question was also the industry’s way of seeing it — more specifically, the interpretive lens through which the event was being understood. In reviewing a range of informative postmortems, we found much of the work highly useful on crypto-native mechanics and local exchange system dynamics, and it served as important literature for our own approach to the subject. At the same time, we also felt there was a missing bridge between the reality of the event and the way it was being represented. Much of the analysis, understandably shaped by the industry’s own conventions, approached crypto as a largely self-contained system.
For us, that lens felt incomplete when it came to certain external shocks. Crypto systems — especially liquidity systems — are not best understood in isolation. They sit within a broader macroeconomic and geopolitical environment, interacting through prevailing macro conditions, market confidence, liquidity provision, and many other forces. Especially in a volatile period of changing world order, macro policy, geopolitics, liquidity, and onchain market structure do not sit in separate boxes, but increasingly collide inside the same system. In that kind of environment, understanding stress is not just about tracking volatility after the fact. It is about seeing how disorder moves, where it enters, under what conditions it enters, how it propagates, and what kind of structure either absorbs it or forces it through.
Our Keynesian Reflex Paper as Foundation, and Strategic Tooling as a Missing Category
It was the need to make a missing interpretive bridge more legible that led us to publish our paper, The Keynesian Reflex vs. the Invisible Hand: Market Structure Lessons from the $19B Flash Crash. The paper was an effort to map cross-domain crisis-response regimes, examine the limits of a liquidation-only market structure, and introduce a shared language for comparing stabilization designs across venues and ecosystems before the next storm. Its goal was not simply to argue for one model over another, but to make tradeoffs around shock absorption, coordination, and market structure more comparable, discussable, and transparent for market-structure stewards.
At one level, the paper was an attempt to think more clearly about liquidity, coordination, and control under unstable exogenous conditions. But in the process, it became something more. It gave public form to a particular interpretive lens: a way of reading the relationship between macro disorder and crypto’s internal plumbing. It took a fast-moving and chaotic set of forces and compressed them into a transmission framework that decision-makers could actually reason with. What had felt diffuse became more legible. What had looked like noise began to take on form.
That was a breakthrough moment for us, because what makes cross-domain shocks so difficult is not only their violence, but the challenge of parsing the many moving parts inside them — policy shifts, macro conditions, geopolitical developments, risk appetite, and onchain liquidity mechanics — while the event is still unfolding. Too many things move at once. Some developments are noise, some are regime shifts, and some begin far away from the place where their consequences are eventually felt. In that kind of environment, clarity is part of the decision itself. Looking back, that was our first real encounter with what we now call Strategic Tooling: a way of compressing raw complexity into legible, modular frameworks that decision-makers can actually work with. There is abundant tooling for developers building blockchain systems and applications. There is far less tooling for helping decision-makers see what they are really deciding about when billions of dollars and millions of users are exposed to cross-system volatility.
Systematized Interpretive Layer for Macro-to-Onchain Transmission and AI-native Architecture as Natural Progression
The conviction that this interpretive lens needed to be systematized came from both market feedback and first-principles reasoning. Following the flash crash, we received genuine curiosity and valuable feedback, and began to see real resonance with market structure stewards. Our Keynesian Reflex paper also sparked active discussion and reasoning, as decision-makers began to reason through the shock within that frame and account for boundary conditions beyond the crypto-native system. As we kept working through the implications of such interpretive frame with constructive market feedback, a larger realization began to take shape. The gap was not only analytical, but operational. Static analysis, however sharp, is not enough in a world shaped by faster-moving signals, tighter coupling, and greater uncertainty.
It became increasingly clear to us that the question now is architectural. Can large amounts of information be parsed, interpreted, structured, and acted upon ahead of time? What kind of system can hold this interpretive work and make it more responsive at scale without flattening what makes it valuable in the first place? The work now seems to ask for a product form: not to replace judgment, but to give it a more structured, legible, and scalable operating layer. From this requirement, an AI-native architecture begins to emerge as a natural progression and continuation of our initial interpretive frame for macro-to-onchain transmission: a better decision layer for moments when macro, policy, and geopolitical developments begin to press on onchain market structure before their consequences are fully visible.
What interests us here is not computation alone but interpretation at scale. A machine does not naturally know which developments matter, which changes are noise versus regime shift, what ambiguity is material, or how to reason through signaling, incentives, and strategic behavior — the kinds of conditions that do not sit neatly inside deductive logic. Before anything can be structured, something has to be seen, sorted, and interpreted. In the age of AI, as the reorganization of the cognitive assembly line lifts the bottleneck of intelligence, much as the First Industrial Revolution reorganized the physical assembly line, intelligent machines can for the first time reason not only through deductive inputs, but also through strategic, qualitative, and broader structural ones. This is why the interpretive layer is the core part of our system. It is the system’s point of view on reality: the layer where messy events are framed, relevance is determined, and complexity is turned into something decision-useful.
If we recall our founding conviction — that “systems encode values, assumptions, and a worldview about how the world is organized” — then the question of how systems behave under stress is never purely technical, as shown in the $19B flash crash. In an era of changing world order, people and institutions live downstream of the systems they depend on. When those systems are not positioned to buffer stress early, the consequences are not abstract: disorder passes downstream into basic economic functioning and the continuity of life. We believe there are ways to contribute to infrastructure continuity and resilience, so that the systems people rely on can better sustain basic functioning and absorb disruption under stress.
We began by trying to understand how systems behave under stress. We are now building toward an Ai-native Interpretive Layer for Macro-to-Onchain Transmission at scale to help market structure stewards interpret and prepare for those stresses more effectively. The framework came first. The interpretive layer follows naturally from it.
In a world defined by volatility, fragmentation, and faster-moving signals, the challenge calls for better ways to make sense of what matters and sufficiently mitigate before the consequences fully arrive.
That is the problem we care about.
And that is the work we are building toward.