Candidates say they want to be a quant and mean any of three things. The firms know exactly which one they are hiring, the interviews are built accordingly, and turning up with a preparation aimed at the wrong role produces a rejection that reads as a technical failure when it was really a targeting one.
The three, compared
| Researcher | Trader | Developer | |
|---|---|---|---|
| Core question | Is this signal real? | What is this worth right now? | Does this run correctly and fast? |
| Daily work | Data, hypotheses, backtests, papers | Monitoring, risk, execution, intervention | Systems, pipelines, latency, reliability |
| Timescale | Weeks per project | Seconds to hours | Sprints and production incidents |
| Tested on | Statistics, probability, ML, research design | Mental maths, games, expected value, decisions | Algorithms, C++ or Python, systems design |
| Typical background | PhD or masters in a quantitative field | Maths, physics, engineering; olympiad and games | Computer science, competitive programming |
| Failure mode | Elegant research that does not trade | Good instincts with no discipline | Fast code solving the wrong problem |
The researcher
Researchers generate and test hypotheses about what predicts returns. The day is data work, statistical testing, reading, and writing up results that are usually negative. Tolerance for negative results is the trait that distinguishes people who last.
The interview tests statistics far more than mathematics for its own sake: regression and its assumptions, hypothesis testing, time series, cross-validation, and above all the ways a backtest lies. Expect to be asked how you would know whether a result is real, and expect the answer to be the centre of the conversation. Many firms set a take-home research task on a real dataset, marked on process and honesty as much as on the result.
The trader
At a systematic firm the trader is not choosing positions by intuition. They own the live behaviour of strategies: monitoring risk, managing inventory, handling the situations the model was not built for, and deciding when to intervene. At a market-making firm the role is more immediate, involving pricing, hedging and reacting to flow.
The interview is unlike anything else in finance. Mental arithmetic under time pressure, probability puzzles, expected value questions and trading games where you must quote a price, accept the fills and manage the resulting position. The games are not testing arithmetic. They are testing whether you make decisions with incomplete information, update on evidence and stay disciplined when losing.
The developer
Quantitative developers build and run the systems everything else depends on: data infrastructure, backtesting frameworks, execution and order management, risk systems, and at the latency-sensitive firms the low-level code that makes speed possible.
The interview is a strong software engineering interview with a domain accent: algorithms and data structures, concurrency, memory and cache behaviour, systems design and, at high-frequency firms, detailed C++ questions. Knowing finance helps and is not the bar. This is also the most accessible of the three for a candidate without a quantitative postgraduate degree, and it pays comparably.
Which one suits you
Answer honestly rather than aspirationally, because the interviews will find out.
- Do you enjoy spending three weeks on something that probably will not work? Research.
- Do you like deciding quickly with incomplete information, and can you be wrong without it ruining your afternoon? Trading.
- Do you enjoy making systems correct and fast, and does a subtle bug bother you until you find it? Development.
- Do you want the highest ceiling on pay with the highest variance? Trading at a proprietary firm.
- Do you want quantitative work with more predictable hours and a clearer path without a doctorate? Development, or a bank quantitative desk.
Pay and progression
Entry compensation at the top proprietary trading and market-making firms is the highest in finance for a graduate, including for developers, and it is heavily weighted to a bonus that genuinely varies. Systematic hedge funds pay well and typically below the proprietary firms at entry, with more of the upside arriving later.
Progression is flat and fast by the standards of banking. Titles matter little, and what changes is the size of the risk budget a researcher influences, the book a trader runs, or the criticality of the systems a developer owns. Bank quantitative desks pay less than either and offer more predictable hours and a more structured career, which is a legitimate trade rather than a consolation.
Moving between them
Developer to researcher is the most common move and it happens regularly, usually by someone who built the research infrastructure and started using it. Researcher to trader happens at systematic funds where the boundary is already soft. Trader to researcher is rarer, because the statistical depth is hard to acquire alongside a live book.
The practical implication for a candidate who is unsure: a development seat at a good firm is a legitimate route in, and it keeps the other two doors open.
Frequently asked questions
What is the difference between a quant researcher and a quant trader?
A researcher tests whether a predictive signal is real, working in weeks on data and statistics with mostly negative results. A trader owns live behaviour: risk, inventory, execution and intervention, on a timescale of seconds to hours. The interviews are entirely different.
Do you need a PhD to be a quant?
For research seats at many systematic funds it is the norm, though a strong masters with relevant work can substitute. For trading it is not expected at all, and for development it is irrelevant. The doctorate is a proxy for research experience rather than a requirement in itself.
Which quant role is easiest to get into?
Development, by a clear margin. The pipeline is larger, the bar is a strong software engineering interview rather than a research track record, pay is comparable at the top firms, and it is the most common route into research later for people who decide they want it.
What do quant trading games test?
Decision-making under uncertainty rather than arithmetic. Quoting a two-way price, accepting the fills you get, managing the resulting position and updating as information arrives. Firms watch whether you stay disciplined when losing, which is harder to fake than mental maths.
Related guides
Want this applied to your recruiting?
Reading is the easy part. For practitioner feedback tailored to your situation, work 1:1 with Suro.