Evidence-backed discovery
New names pass exchange, price, liquidity, momentum, and relative-strength checks.
Autonomous AI Trading · Fully Auditable
An AI trading desk
that does the research
A language model writes its own market research, discovers stocks, and decides when to buy, sell, or wait—with every decision explained and recorded.
Every decision explained · Fully auditable
SYSTEM.ACTIVE NY --:--:--
01 / Decision makerLanguage ModelResearch written in-house
02 / UniverseUS EquitiesLiquid, screened names
03 / ExecutionPlan → FillDollar-based decisions
04 / AuditEvery DecisionResearch, reason, and fill
Why it works
A / Rules
A / Rules
A Trading Desk With Rules
Clear constraints keep the agent focused on durable evidence instead of short-term noise.
Watch before buying
Discovered names are watched first and must earn a place before they earn capital.
Three trading clocks
Monthly planning, weekly review, and hourly patience work together.
Sell before buy
Released cash can fund the next stronger opportunity; ownership is always checked.
Evidence over noise
One green hour is not enough. Leadership must persist across the week and month.
Transparent record
Research, rationale, amount, and account fill stay reviewable in the dashboard.
Process
S0 → S4
S0 → S4
From Research to Account Fill
A deliberate workflow built to separate signal from noise.
Find candidates
Searches for liquid US common stocks with momentum and strength versus the Nasdaq-100.
Build the plan
Monthly theses set the direction; weekly reviews test each view against evidence.
Choose an action
Buy, sell, or hold with a symbol, dollar amount, and reason grounded in the plan.
Measure the account
Orders fill at the quoted price; equity is cash plus the marked value of shares held.
Profit measurement
The Account Tells the Story
Profit is the change in account value: cash plus the current value of shares still held. Gains are marked while positions are open and realized when sold.
- Cash from a sale funds the next buy
- Persistent laggards can fund stronger leaders
- The desk may stay in cash when nothing is stronger
- Every decision and fill is visible in the dashboard
OrdersDollar Based
FillsQuoted Price
ResultCash + Shares
EquityMarked Account Value
SAMPLE
Trust & risk
Guardrails
Guardrails
Designed to Be Inspected
Automation does not mean a black box.
Transparent by design
Read the research and rationale behind each action.
Rules before decisions
Plans set boundaries before hourly checks can act.
Risk controls
Cash rules, position limits, and ownership checks guide execution.
STATUS / PAPER SIMULATION
For research and educational purposes. Trading involves risk, including loss of principal. Past and simulated performance does not guarantee future results. Nothing here is investment advice.
For research and educational purposes. Trading involves risk, including loss of principal. Past and simulated performance does not guarantee future results. Nothing here is investment advice.
Questions
FAQ
FAQ
Questions, Answered
A clear view of what the agent does—and what it does not do.
It starts with a fixed list of liquid names, then searches for additional US common stocks. Candidates must pass exchange, price, liquidity, momentum, and Nasdaq-100 relative-strength checks.
The displayed combined desk currently represents paper or simulated trading. Orders are filled at quoted prices inside the account simulation.
Account equity equals cash plus the current marked value of shares held. Open gains are unrealized; gains become realized when a position is sold.
Yes. The dashboard records the action, dollar amount, research, strategy rationale, and account fill for review.
Market prices can fall, research can be wrong, and simulated results may not reflect live execution. Automation does not remove investment risk.
Next step
See Every Decision in Context
Review the latest research, positions, fills, signals, and equity curve in the dashboard.
View Live Dashboard →