Quantis AI Trading
An autonomous AI agent trades a US-stock paper account with regime-gated leverage, conviction sizing, and a defined-risk options sleeve — every position wrapped in hard survival mechanics the LLM cannot override.
Personal engineering project · Paper trading only · Not investment advice
An aggressive, regime-aware trader that scales risk up in favourable markets and pulls it back hard when conditions turn. An LLM reads the market, sets a leverage budget, sizes by conviction, and expresses ideas through equities, a leveraged index core, and a defined-risk options sleeve. The leverage is the offence; the drawdown ladder, VIX kill-switch and margin-distance gate are the defence. Survival is rule one — a wipeout can't compound.
How it works
Read the regime
An Opus pre-market routine reads market context — SPY vs its 50/200-day MAs, VIX, futures, catalysts, earnings, sector momentum — and sets today's regime, gross-exposure ceiling, and the profit-ratchet cap (max leverage steps down as the account makes new highs, banking gains). Strong uptrend earns leverage; risk-off forces it back to cash.
Deploy with conviction
Re-validates plans on live quotes, then sizes by conviction (high / medium / standard) within the regime's leverage budget. Single names must clear a trend-confirmation gate (above their 20-day MA, positive momentum) — strength only, no falling knives. A leveraged QQQ/SPY core absorbs idle cash and gets a catastrophic backstop stop; every discretionary fill gets a volatility-adaptive ATR stop.
Cut fast, let winners run, survive
Two extra intraday risk-checks (10:30 + 13:30 CT) plus midday hunt for trouble. Losers are cut fast at −1.5×ATR or −7%; winners move to breakeven, bank a third at +1.5×ATR, then ride a trailing Chandelier with no fixed cap. Survival mechanics run continuously — drawdown ladder (−10/−20/−25%), VIX kill-switch, margin-distance gate, a fast-reversal trigger, and an into-the-close de-lever. Friday's Opus review grades and runs the kill-check.
How it's built
No databases, no ORM, no in-memory state. Every memory file — strategy rulebook, trade ledger, daily research, the atomic risk-state file, weekly reviews — is committed to a private GitHub repo. Every routine run is a fresh container that clones, decides, commits, exits.
Hard rules the LLM cannot override
Aggression is allowed only inside a cage of survival mechanics enforced before any order is placed. The LLM sets the regime and picks the trades; the rules cap the leverage, size the options sleeve, and pull risk the moment drawdown or volatility crosses a line.
- Regime-gated leverage — hard cap 3.0× gross, earned only in strong uptrends
- Conviction sizing: high 35% / medium 20% / standard 12% of equity
- Trend-confirmed entries only — above the 20-day MA with positive momentum
- Cut losers fast — exit at −1.5×ATR or −7%, whichever hits first
- Let winners run — breakeven move, scale out a third at +1.5×ATR, then a Chandelier trail with no fixed cap
- Profit-ratchet — max leverage steps down as the account makes new highs, banking gains
- Catastrophic stop on the leveraged core; volatility-adaptive ATR stops on every single name
- Drawdown ladder: −10% halve · −20% go flat · −25% hard halt
- VIX kill-switch, 15% margin-distance gate, and a same-day fast-reversal trigger
- De-lever into the close + weekends; defined-risk options only (premium = max loss, ≤15% of NAV)
What this actually is
This is a learning project. The infrastructure — an autonomous LLM agent, hard-rule risk discipline, git-as-memory, an atomic risk-state file — has engineering value regardless of P&L. There is no signal service, no subscription, nothing for sale, no community — just a private dashboard for the operator and a public page explaining what was built.
See it run
The dashboard is private. The only person with access is the operator. Click below if that's you.
Sign in