Σ
Adaptive Alpha
Long- and short-horizon signal systems
Rolling models read recent regimes instead of treating market structure as static. Multiple mechanisms, one rigorous evidence language.
Quantitative research / Hong Kong / US
Markets areliving systems.
Dnalyaw is an AI-native quantitative trading lab focused on US and Hong Kong markets. We bring signal research, execution, and independent risk controls into one continuous validation process.
Component
Component
Component
Signal intelligence
Adaptive, long- and short-horizon, multi-mechanism
Execution science
Costs and fill conditions in the research loop
Machine governance
Risk veto remains absolute
Evidence memory
Link research, risk, and execution
Capabilities & research directions
Multiple mechanisms.Shared research discipline.
We study signals across horizons and markets, with trading costs and independent risk controls built into the validation process.
↝
Execution Science
Research that survives contact
Signals are studied against the prices, frictions, timing, and failure modes they will face in the market — not in an idealized vacuum.
N
Portfolio Intelligence
Multi-strategy portfolio research
We study strategies with different mechanisms and test their correlations and shared failure modes under stress.
ψ
Neural Market Models
Adaptive representations of market state
Dynamic and recurrent models study how useful state can persist, update, and reset across changing regimes without treating market structure as stationary.
π
Adaptive Execution
Machine-learned execution policies
A broader execution intelligence program spanning supervised models, online adaptation, and reinforcement learning for placement, timing, and capital efficiency.
⌁
Market Microstructure
US–HK market microstructure
Research into how sessions, venues, and latency affect execution quality. Co-location remains a frontier research direction.
Frontier programs describe research directions, not capabilities admitted to live trading.
Our evidence method
Evidence beforeoutcomes.
Define the hypothesis and evaluation criteria before observing outcomes. State the sample, cost assumptions, and limits alongside each conclusion.
Record the question first
Specify the hypothesis, data scope, and evaluation criteria before observing results.
Challenge the explanation
Check point-in-time data, trading costs, and stress conditions. Distinguish replay, forward observation, and live execution.
Retain failures and limits
Keep rejected research and its boundaries so later work can be explained and reviewed.
How does risk remain independent from research?
This engineering article describes research producing target portfolios, execution managing orders, and independent risk checks on trading requests. It also discusses checking partial fills and rejections in simulation.
The article documents the design at that time. It does not establish current operating status, audited performance, or strategy effectiveness.
Read the case and design tradeoffsSignals learn.Risk does not deform.
Each step from research output to trading request has a clear owner. Independent risk controls determine whether requests proceed; execution records inform the next research cycle.
System flow illustration
Shows responsibilities and information flow. No live market or account connection.
Research
Propose and test candidate signals
Portfolio
Translate signals into target exposure
Independent risk
Check, reduce, or reject requests
Execution
Manage orders and record outcomes
Execution outcomes and costs → next research cycle
Research signals have no authority to bypass risk controls.
Execution feedback
Fill conditions
Compare expected prices, actual fills, and waiting time.
Costs and tails
Observe slippage, impact, and exceptional conditions.
Decision traces
Link research output, risk decisions, and execution records.
Forward ↔ Inverse
The outcome is visible.The mechanism is not.
Similar price paths can come from different market mechanisms. Mathematics helps us propose explanations; point-in-time data, trading costs, and forward tests determine whether they are useful.
Explore the mathematics and validation methodsFounder & selected writing
Research throughbuilding.
Wayland Zhang, founder of Dnalyaw, is an AI researcher and systems builder. His writing spans neural networks, AI agent architecture, and quantitative trading.
Meet the founder ↗Mathematical foundationA Question Chern Asked, Answered by Two Worlds Numerical manifolds and neural networks arrive at the same answer, thirty-seven years apart.Neural-network foundationThe Four Realms of Neural Networks From PDE solvers to manifold geometry and measurement — a framework for thinking about learned representations.Quantitative researchThe Geometry of Alpha Why the durable quant moat is a factory that keeps producing and validating signals, not a single strategy recipe.Research systemsDnalyaw, Engineered from Scratch From research output to execution feedback: system responsibilities and risk boundaries.