1:04Transcript
Claude Trading Strategies: I Tested SPY Timing Against Buy-and-Hold
Connect Claude to NexusTrade with MCP and compare an SPY 200-day moving-average backtest with buy-and-hold, including returns, drawdown and strategy rules.
NEXUSTRADE / VIDEOS
AI trading videos
Shorts I filmed while building NexusTrade.
Every video has a transcript.
91 shorts · every one transcribed
Algorithmic trading videos
Start watching ↗1:04Transcript
2:17Transcript
AI High-Frequency Trading? This 100ms Claim Doesn’t Hold Up
Austin Starks fact-checks Ray Fu’s Jev AI trading pitch, challenging its 100ms HFT claim and asking for out-of-sample tests and evidence of better results.
1:17Transcript
Congressional Trading Disclosures. Two Free Repos to Research Them.
Download public congressional trading disclosures and test ideas with a free Python SDK. Filings are delayed, so use the public disclosure date.
1:11Transcript
How to Read a Flamegraph: I Cut My Backtest Time in Half
Reading a real flamegraph from a Rust backtesting engine: the fixes that halved a ten-minute options backtest, and the parallel loop that made it 73% slower.
1:25Transcript
Create a Trading Strategy with Cursor. Then Deploy It to Alpaca.
Create a trading strategy in Cursor and deploy it to Alpaca through the NexusTrade MCP. The $13,805 book on screen is his existing Alpaca account.
1:39Transcript
How My AI Trading Bot Decides What to Buy and Sell
The three steps an AI trading agent takes to pick trades: writing a deterministic rule, walk-forward testing it over four folds, and deploying it to paper or live.
1:22Transcript
Claude Code Trading Strategy: 1,100% in 11 Years, and a 59% Drawdown
One command connected Claude Code to my backtest data. The rule it kept turned $10,000 into $119,942 over 11 years, with a 59% worst drawdown.
1:41Transcript
I fact-checked a day trading strategy with 423,000 views. It loses to doing nothing.
A day trading strategy with 423,900 views, coded by an AI agent and backtested on a year of minute data. Tuned and charged zero fees, it still lost 2.40% out of sample while the index gained 1.02%.
1:26Transcript
Why Is My Backtest So Slow? GPT-6 Astra Cut Mine to 51 Seconds.
My backtests took ten minutes. GPT-6 Astra profiled the Rust engine, cited a 2014 paper, and got the run down to 51 seconds on hardware I already owned.
1:24Transcript
Every Rule My AI Trading Bot Follows Is Public. So Is the 12% Crash.
Every rule, backtest, crash and bug behind my $25,000 AI trading account is public. Here is what the live book actually did.
1:13Transcript
Parquet for Backtesting: My $230 Disk Bill Is Now $34 a Month
How Parquet chunk statistics speed up options backtests, and how one copy on Tigris replaced seven duplicate disks and cut a $230 monthly bill to $34.
1:04Transcript
What Is Lookahead Bias? Why Your Vectorized Backtest Is Lying
Why a vectorized Python backtest can show a fake equity curve: lookahead bias, position-dependent rules, and how an event-driven backtester avoids them.
0:14Transcript
AI Data Cleaning: From SEC Filings to Checked Trading Data
See how an LLM, SEC filing retrieval and repair code work together to clean trading data, check results and retry in an agentic workflow before backtesting.