This is how most HFT firms actually make money. Think of it like being a very fast used car dealer.
You simultaneously offer to buy Apple stock at $150.00 and sell it at $150.01. If someone hits both sides, you make $0.01 per share. Doesn't sound like much, right?
Now multiply that by millions of shares per day. Suddenly you're making serious money. Jane Street and Citadel Securities became huge doing exactly this.
Speed Arbitrage
Same stock trades on multiple exchanges. Sometimes the price on NYSE is $50.00 while Nasdaq shows $50.01. HFT firms with faster connections buy at $50.00 and sell at $50.01, capturing the difference before regular traders even see the opportunity.
Statistical Arbitrage
When an ETF that tracks the S&P 500 gets slightly out of sync with the actual S&P 500 stocks, HFT algorithms notice immediately and trade to capture the difference.
The Technology Race
HFT firms spend millions on infrastructure most people don't even know exists:
Co-location: They literally put their servers in the same building as stock exchanges. Physical distance matters when you're racing light signals.
Microwave towers: Some firms use microwave and shortwave radio to transmit data faster than fiber optic cables.
GPS clocks: They need nanosecond-level precision timing.
The goal isn't to be smart. It's to be first.
What This Means for Your Interview
If you're applying to Jane Street, Citadel Securities, HRT, Virtu, or Jump Trading, you're interviewing at an HFT firm. Even if they don't mention trading in your job description.
They don't ask "How would you build a latency arbitrage algorithm?" Instead, they test the underlying skills:
Mental math under pressure: Because their systems process thousands of price changes per second, they need people who can think clearly when things get hectic.
Probability and stats: Every HFT strategy is basically a probability model. They need to know you understand expected value, distributions, and risk.
Problem-solving speed: Not because you'll be manually trading, but because you'll be building and debugging systems that need to react in microseconds.
The Interview Reality
Most HFT interview questions look like pure math puzzles. "You have three dice, what's the probability of rolling the highest number?" or "How would you price this simple betting game?"
They're not testing your knowledge of trading. They're testing whether your brain works fast enough and accurately enough to eventually build the models that power their systems.
Why It Matters for Markets
Love it or hate it, HFT has fundamentally changed how markets work. Bid-ask spreads are much tighter now. It's much easier for regular investors to buy and sell stocks at fair prices.
But it's also created a world where having the fastest connection matters more than having the best analysis. Traditional "fundamental" investing looks pretty slow when machines are making millions of decisions per second.