QuantMechanix: An options research desk, built in-house.
Each market day morning it scans 104 stocks and ETFs, runs statistical models and writes a ranked brief. Written risk rules can check any idea.
Built for our own research. It places no orders and is not a claim of investment performance.

One market day morning
The ideas fall as you scroll.
Select any stage below to view its details.
- Getting the right data in, clean and connected.
- Statistical models, and a running record of whether they work.
- Making it run by itself, and recover when a step fails.
Scan
Scan: Before the market opens, a fixed list of 104 stocks and ETFs is scanned.
-
UniverseThe same written list of 104 stocks and ETFs every market day.
-
InputsNo model reads news, options flow or earnings dates.
Research data never sets a score or a direction, and a failed symbol is a recorded gap.
Criteria
Criteria: A first screen on liquidity, volatility rank and earnings dates.
-
LiquidityThinly traded symbols are set aside.
-
Volatility rankKeeps implied-volatility rank above 20.
-
Earnings datesbuilt, not switched onSets aside symbols within 2 days of earnings; date refresh is off.
At most 60 of the 104 go on to the models.
Models
Models: Fourteen models run every market day, in families. Point at one for its equation.
Nine models vote on direction, in up to six families
-
Industry standard
Hidden market-state detector
Three-state hidden Markov model, market mood family.
Industry standard
-
Industry standard
Price mean reversion
Ornstein-Uhlenbeck on log price, snap-back family.
Industry standard
-
Industry standard
Jump detector
Merton jump-diffusion, shock family.
Industry standard
-
Industry standard
Sell-off cluster detector
Hawkes self-exciting process, shock family; can only vote up.
Industry standard
-
Industry standard
Moving-average stack
20 and 50-day moving averages, trend family.
Industry standard
-
Industry standard
Channel breakout
20-day Donchian channel, trend family.
Industry standard
-
Industry standard
Rate of change
10-day rate of change with a 14-day RSI, trend family.
Industry standard
-
Industry standard
Composite scorer
Registry-weighted blend, composite family; scores -50 to +50, votes past 10 either way.
Industry standard
-
Experimental
Path-integral tilt
Feynman-Kac terminal-density asymmetry, its own family; low weight, capped confidence.
Experimental
Five models report on volatility and risk
-
Industry standard
Volatility mean reversion
Ornstein-Uhlenbeck on realized volatility; five-day forecast.
Industry standard
-
Industry standard
Stochastic-volatility forecast
Heston variance dynamics, fixed parameters; ten-day forecast.
Industry standard
-
Industry standard
Volatility clustering
GARCH(1,1), normal innovations; five-day forecast.
Industry standard
-
Industry standard
Return disorder
Normalized Shannon entropy over 20 bins; labels disorder high, moderate or low.
Industry standard
-
Industry standard
Crash-together risk
Clayton copula against the S&P 500 ETF; labels crash-together risk high, moderate or low.
Industry standard
Evidence
-
Skill scoringThe five risk models are never scored.
One vote per family. A pass needs strength of at least 0.60 and three agreeing families.
Evidence ladder: Unproven under 30 settled predictions, Earning from 30, Negative from 50 with no skill, Earned from 200 with skill.
One label covers the set. Labels change nothing by default; promotion is manual.
Market conditions
Market conditions: 29 written risk rules can check an idea, on request and on orders. They are not part of the scheduled morning run.
-
Pre-trade checklistRun on request and on option buy orders; every verdict is written down.
-
Blockbuilt, not switched onCan stop an order; the order path fails closed.
-
Paper by defaultNo scheduled job places an order; going live needs a typed confirmation.
-
Live executionplannedGated live trading and an autonomous paper trader are later phases; not built.
The rules record; they do not block.
Brief
Brief: The same-day review candidates, ranked. Direction comes from the composite scorer, one of the 14.
-
DeliveryA script renders a PDF brief and can email it; skipped when no address is set.
Each idea's chance of clearing breakeven by expiry averages a skew-implied density, Monte Carlo and a Markov regime model.
Data engineering
Getting the right data in, clean and connected.
Every scan price comes from one brokerage connection.
When data looks wrong, the run is flagged, not mistaken for a quiet day.
Yours could be orders, payments or inventory.
- 104stocks and ETFs scanned
- 1source for the scan's live prices
- 4self-checks for bad data
Engineering detail: data engineering
Stack
- Python 3.11
- FastAPI
- Schwab API, via schwab-py
- PostgreSQL
- pgvector
- SQLAlchemy
- Alembic
- Redis
- pandas
- NumPy
- SciPy
- statsmodels
- APScheduler
Decisions
- Research data never sets a score or a direction.
Where each part stands
- Quote, price-history and option-chain calls paced
- Its own daily implied-volatility history
- An earnings calendar
Models and evaluation
Statistical models, and a running record of whether they work.
Fourteen models study each shortlisted stock: nine say up, down or neither, and five say how rough the ride may be.
Votes are scored later against what really happened, and no model is called proven before the results are in.
Yours could be an AI feature scored against real outcomes.
- 14models run each market day
- 30scored predictions before any rating
- 200scored predictions for the top rating
Engineering detail: models and evaluation
Stack
- NumPy
- SciPy
- statsmodels
- scikit-learn
- in-house walk-forward backtester
- in-house discovery harness
- Postgres
- pytest
Decisions
- Labels change nothing by default; promotion is manual.
Where each part stands
- Fourteen models run daily: nine vote, five report risk
- An evidence ladder: Unproven under 30, Negative from 50, Earned from 200
Systems engineering
Making it run by itself, and recover when a step fails.
It starts at 5:30 every market day morning and saves its progress as it goes.
A passing failure, like a network error, is retried automatically; a stopped run can pick up where it saved.
Yours could be a nightly report that finishes while you sleep.
- 2automatic retries per step
- 3save points in every run
- 29checks recorded on each proposed trade
Engineering detail: systems engineering
Stack
- Docker Compose
- Ubuntu
- APScheduler
- Postgres
- Alembic
- Redis
- FastAPI
- React
- Vite
- TypeScript
- ReportLab
- Matplotlib
- smtplib
Decisions
- Risk rules record every verdict; enforcement is opt-in.
Where each part stands
- A checkpoint after each tier; a stopped run can resume from it
- 29 written risk rules, every verdict recorded
- A confirm pass that re-prices after the open
What this means for your system
A Production Triage reads your system from the code and, in three business days, hands you a ranked risk list and a fix plan.
See all pricesStart with fifteen minutes.
A free call about your system. We'll tell you which engagement fits, or that none does.