Research
The work begins with reinforcement learning and expands outward into market systems, safety, portfolio cognition, and machine economies.
Themes
RL behavior
Studying how agents preserve proxy rituals after reward landscapes shift, and what that reveals about adaptation, credit assignment, and machine belief.
Proxy persistence, delayed credit, adaptation failure
Safety
Investigating how systems stabilize around patterns that are not causally grounded, then designing interventions that make agent behavior legible and corrigible.
Non-causal loops, interpretability, intervention design
Markets
Architectures for agents that reason over market state, risk, execution, regime change, and the difference between signal discovery and overfit mythology.
Execution intelligence, risk memory, market adaptation
Financial systems
Decision infrastructure for portfolios that combines probabilistic views, constraints, capital allocation, and explainable operating discipline.
Allocation engines, scenario logic, capital cognition
Coordination
Protocols for funding, attribution, trust, and coordination where software agents and human communities transact around shared economic intent.
Agent economies, incentives, decentralized labor
Cognition
Exploring how identity, context, memory, and feedback shape decisions across humans and agents operating under uncertainty.
Adaptive cognition, identity models, decision loops
Contact
Bring the hard shape of the problem: ambiguous incentives, moving data, uncertain rewards, brittle workflows, or research that needs to become a product.