What do quant traders and quant researchers do all day?

What does a quant trader's day look like, hour by hour?

This is a typical day for a trader running a market making strategy in US equities or options at a prop shop. Firms differ in the details, but the shape is common: preparation before the open, attention during the busy parts of the session, and analysis after the close.

• Before the open: read overnight moves, check scheduled events (earnings, data releases, index changes), and set or confirm strategy parameters for the day.
• The open: the busiest stretch. Watch fills, positions, and risk limits, and step in when the market moves faster than the model expects.
• Late morning: quieter. Look into anything odd from the open, talk with researchers about a parameter that misbehaved.
• Midday: research and tooling time for traders who also build. Many firms expect traders to code.
• Into the close: volume picks up again. Manage end of day positions and risk.
• After the close: review P&L by strategy and by trade type, write up what went wrong, and plan changes for tomorrow.

How does a market maker actually make money? A worked example

A market maker earns a small edge on a large number of trades and loses when it trades with someone who knows more. Assumptions: a stock's fair value is 100.00 and the firm quotes a bid of 99.98 and an offer of 100.02. Fees and rebates are ignored.

1. Each fill earns about 0.02 against fair value: buying at 99.98 or selling at 100.02.
2. If 40,000 shares trade against uninformed flow during the day, that is about 40,000 x 0.02 = 800 dollars of edge.
3. Now one informed buyer lifts the offer for 10,000 shares at 100.02, and the price moves to 100.10. The firm is short 10,000 shares that are now worth 0.08 more each.
4. Loss on that trade: 10,000 x 0.08 = 800 dollars. One informed trade erased the whole day's edge.
5. So the trader's real job is adverse selection: noticing when the flow is informed and widening, shrinking, or pulling quotes before it costs money.

What does a quant researcher's day look like?

A researcher's day has less adrenaline and more depth. There is no single opening bell for research; the rhythm is set by the projects, and a single idea can take weeks. A typical day looks like this.

• Morning: check how yesterday's signals performed live versus what the backtest said they would do.
• Most of the day: write code to clean a dataset, build a feature, and run a backtest.
• Review: present a result to colleagues, who try to break it.
• Afternoon: read a paper or an internal write-up, then decide whether to push an idea further or kill it.
• Ongoing: work with engineers to move a signal that survives review into production.

Why do most research ideas fail? A worked example

Because noise looks like signal if you test enough ideas. Assumptions: you test 100 signals that in truth predict nothing, each with a standard test at the 5 percent significance level.

1. By chance alone, you should expect about 100 x 0.05 = 5 of them to look significant.
2. If you keep those 5 and drop the rest, your results will look great in the backtest and fail live.
3. Researchers defend against this by holding out data they never touch until the end, demanding an economic reason for each signal, and adjusting for how many tests they ran.
4. This is why interviewers ask about overfitting and look-ahead bias. They want to know you would catch your own false positive.

Which quant programs are open right now?

The table below reads each firm's own careers site. Quant recruiting runs early: the path window is Aug to Oct, fills by Dec, and trader, researcher, and developer internships are separate applications at most firms. Pick the role whose day above sounds like yours.

What do quant interns actually do over a summer?

Intern programs at prop shops and quant funds commonly mix classes with real projects, though the details are each firm's own. Trading interns often start with education on markets and the firm's tools, then play mock trading games, then work on a project with a trading desk. Research interns usually own one research question for the summer and present the result. Developer interns ship code to real systems.

In every case, the evaluation is your work and how you reason about it. Networking matters little in this path; grades, competitions, and what you have built matter more.

• Trading: mock trading, estimation games, a desk project, and reviews of how you size risk.
• Research: one research project with a mentor, a final presentation, and a lot of code.
• Development: production code on a real system, code reviews, and a final demo.

Questions students ask

Do quant traders need to code?

Increasingly yes. Many firms expect traders to write scripts and analysis in Python or similar, even if researchers and developers build the core systems.

Are quant hours better than banking?

Generally yes. The path page puts them at 50 to 60 a week, with the day tied to market hours rather than deal deadlines.

Is quant research only for PhDs?

Many full-time researchers have graduate degrees, but undergraduate research internships exist at a number of firms. Strong math, statistics, and code matter more than the degree label.

What is adverse selection?

Trading with someone who knows more than you. A market maker earns a small spread from most traders and loses to the few who are informed, so spotting informed flow is much of the job.

Do quant firms care about finance knowledge?

Less than banks do. They care about probability, statistics, expected value, and code. Market knowledge helps but can be taught on the job.