The idea
Buy long-dated calls on my watchlist names that are trending up.
Write it the way you would say it. Aurora, the research agent, turns it into rules and backtests them after fees. Then it tests them on later dates.
36,148 people have run 652,784 backtests on NexusTrade. I also trade it with $25,000 of my own money, in public.
Where would you like to start?Let’s build your first plan.Each 252-day test started with a fresh $25,000, and the line chains the three of them. The strategy beat SPY in all three windows it had never seen. Watch the rules change or open the study.
Before each 252-day window, Aurora tested four position sizes on the dates before it and kept the one with the best Sortino ratio. That version then ran on dates it had never seen.
Chosen for window one
Chosen Changed since the last window
Each window’s settings came from the dates before it. The test result never fed back into the choice.
| Window | Test dates | Chosen settings | Validation return | Test return | SPY | Closed trades | Fees | Largest fall |
|---|---|---|---|---|---|---|---|---|
| 1 | Aug 17, 2024 – Apr 25, 2025 | 10% per position | +121.8% | +38.6% | −0.7% | 282 | $239.34 | −31.1% |
| 2 | Apr 26, 2025 – Jan 2, 2026 | 5% per position | +248.6% | +144.0% | +25.1% | 349 | $244.14 | −13.6% |
| 3 | Jan 3 – Sep 11, 2026 | 3% per position | +306.2% | +77.8% | +11.7% | 283 | $123.93 | −10.7% |
Fresh $25,000 per window, fees of 0.1% of each stock trade and $0.65 per option contract, a 14-day embargo before each validation period, and the highest validation Sortino ratio picked among versions that traded at least nine names. The setting searched is the share of the portfolio in each position. These are historical simulations. The live account below runs the current version of the same strategy. Open the study.
Six real NexusTrade backtests, each against SPY, Oct 2021 – Sep 2026. Strategy SPY
Technical indicators
Fundamentals
Economic data
Congressional disclosures
Your own data
All of them together
Why the test matters
Backtested from Jan 2024, the disclosure rules returned +136.0% against SPY’s +68.7%. Run on five windows its settings never saw, the same rules trailed SPY.
Algorithmic Trading Fundamentals
It needs no brokerage account.
Automation runs only after you approve it and switch it on.
NexusTrade
pip install nexustrade
npm install nexustradeYou can also connect the MCP server or the REST API. Read the developer docs.
I’m Austin, the founder. I started this portfolio with $25,000 of my own money on May 5, and every position and fill is public.
So far I’m 6.2 points ahead of SPY.
Follow the Public Portfolio Challenge series Inspect the portfolio and its fills
Fork one and test it yourself. Browse every template.
| Template | Last 12 months | 1-year return | Largest fall | Sortino | Access |
|---|---|---|---|---|---|
| Venezuela Equal WeightBy Austin · Forward test · paper | +53.7% | −15.5% | 3.08 | Free to fork | |
| SPXL Scaling into Drawdowns V5 (Shorter Cooldown)By Austin · Forward test · paper | +52.9% | −12.5% | 2.31 | Free to fork | |
| Drawdown-Based AccumulationBy Austin · Forward test · paper | +39.7% | −32.8% | 1.32 | $39.99 a month |
Each record is labeled paper or live. Updated Oct 21, 2026. Past performance does not guarantee future results.
You did an excellent job of considering all aspects of trading and research which made the application smartly transition from research to strategy, to watchlist, to trade… very nicely done.
Spencer K. Chief Finance Officer / Institutional InvestorI discovered NexusTrade on YouTube and was so impressed that I invited the founder to speak at my MBA class.
Michael P. Senior Lecturer, Harvard Business School
There are not a lot of people (particularly in retail) building what he’s building, and the platform is incredible.
Trevor R. Managing Director at a quant firm
Even though I know how to code, it’s still very enjoyable to be able to do this with such ease. I was able to optimize my strategy to the fullest.
Delanoe P. Artificial intelligence engineer
The founder’s Public Portfolio Challenge uses real capital. Template records are labeled paper or live. The Pelosi backtest and the walk-forward study are historical simulations. In the walk-forward study, each window’s settings were picked using only earlier data, and every date had already passed when the study ran. Open each record to inspect its rules, dates, fees and risks.
No. The three-question start asks about your experience, priorities and interests, then suggests ideas to test. Choose one or bring your own. Aurora works in plain English, and you can also build rules by hand.
Yes, if you choose to. Paper trading needs no brokerage. Live trading connects to Alpaca, Public, TradeStation, Tradier or Robinhood, and stocks, options and crypto are supported. Your brokerage decides which of those you can trade. Automated live trading requires approval, current consent and automation you switch on yourself.
You can fork strategy templates and build strategies by hand for free, with two paper portfolios of up to ten strategies each. Aurora research uses tokens. Paid plans include a daily allowance, and you can also buy tokens without changing plans. How the allowance works.
The monthly figure is the annual charge divided by twelve, so you can compare plans. The Data-Driven Investor plan charges $480 once a year, which shows as $40 a month. You can also pay $49.99 monthly or $13.99 weekly.
Some strategy templates are free to fork, and others have their own subscription price. A platform plan covers its research allowance and platform limits. A paid template subscription covers that one template and leaves your platform plan as it is.