If you are choosing between the three quant seats, ignore the base salary: at the top of the market it is the same number. Jane Street's New York postings list a $300,000 base for its trader, researcher, and software-engineer roles alike (Quant Blueprint, checked Sept 28, 2026). The roles split on four other things: how volatile your bonus is, how high the ceiling goes, which degree gets you through the door, and what the interview actually tests.
The short version: traders own the steepest pay curve and the hardest speed screen, researchers sit behind a PhD gate at many funds, and developers trade a lower ceiling for the most portable skill set. Pick the one whose interview you can win and whose bad year you can survive.
The 30-second verdict
| Quant trader | Quant researcher | Quant developer | |
|---|---|---|---|
| What you do | Run live risk, price and hedge positions in real time | Find and validate signals, build predictive models | Build the trading, data, and low-latency systems |
| Typical degree | Bachelor's or master's (Quant Blueprint) | Often PhD; Citadel's 2027 posting requires one | BS/MS in CS or engineering (Quant Blueprint) |
| Graduate total comp, US | $200K-$500K (Quantt, Sep 2026) | $250K-$475K (Quantt) | $150K-$300K (Quantt) |
| Mid-level (3-5 yrs) | $400K-$900K | $400K-$900K | $300K-$550K |
| Senior (6-10 yrs) | $700K-$2M+ | $700K-$2M+ | $500K-$900K |
| Top of the ladder | PM: $1M-$10M+ | Lead/PM: $1.5M-$5M+ | Staff/principal: $700K-$1.5M+ |
| Interview gate | Mental math, probability, betting and market-making games | Statistics, ML, research methodology | Algorithms, C++, systems, concurrency |
| Pay volatility | Highest: tied directly to P&L | High: tied to strategy performance | Lowest: closer to big-tech curves |
| Exit market | Narrow: other trading firms | Medium: funds, some ML labs | Broad: most of the software industry |
Every comp figure in the table is Quantt's estimate from reported data (US guide, updated Sept 24, 2026), not a firm disclosure. Use the ranges to size the gaps between roles, not to value a specific offer.
Pay: same base, different curves
The posted base is a weak signal because the best-paying firms flatten it. Besides Jane Street's flat $300K, Quant Blueprint's Sept 28, 2026 snapshot of employer postings shows D.E. Shaw paying its New York quant analysts $275,000 with a bachelor's or master's and $300,000 with a PhD, Citadel and Citadel Securities posting $235,000-$300,000 for PhD quantitative researchers, and Optiver posting $200,000 for Chicago traders and researchers. Notice what differentiates base at D.E. Shaw: the degree, not the seat.
The real difference is the bonus line. Quantt's US guide describes the trader curve as the steepest, because traders sit closest to P&L, with portfolio managers ranging from $1M to $10M+ and "no practical ceiling" in strong years. The same guide notes the flip side: that number can fall hard in a losing year. Its researcher guide (updated Aug 6, 2026) gives a cleaner side-by-side at five to eight years in New York: traders $400K-$1M+, researchers $350K-$800K, developers $280K-$500K. The stated reason for the developer discount is attribution: it is hard to pin a slice of revenue on the person who built the order gateway.
Year one at a top firm compresses some of this. Quantt's firm-level graduate estimates put a Jane Street or Citadel Securities trader at $400K-$700K, a Hudson River Trading developer at $300K-$500K, and a Citadel hedge-fund researcher at $275K-$475K. A strong developer at a top market maker can out-earn a researcher at a fund in year one. The ranking re-asserts itself around year five, when the trader's and researcher's bonuses start tracking their own P&L and the developer's does not.
One more wrinkle for researchers: Quantt reports that junior PhD researchers start $30K-$80K higher in total comp than master's hires, a premium that narrows as track record takes over.
Interviews: three different exams
Traders are screened for speed under pressure. Optiver's graduate trader pipeline is a good template: online assessments, an HR screen, a technical interview, and a superday with trading simulations (EverythingQuant). The online stage alone includes the 80-in-8 test (80 arithmetic questions in 8 minutes), NumberLogic (26 sequence questions in 25 minutes), Beat the Odds (10 probability questions), and nine Zap-N reaction games. Later rounds add brainteasers, expected-value betting, and market-making games. Coding is rarely tested. If you freeze on mental arithmetic, no amount of LeetCode will save this loop.
Researchers are screened for rigor. The interview centres on probability, statistics, and model design, often as open-ended questions with no clean answer (Know-Quant, Jan 2026). At PhD-gated funds the CV is part of the exam: Citadel's posting for 2027 asks for independent research experience and the ability to translate models into Python, R, or C++. Expect to defend your own research as hard as you solve their problems. Jane Street's researcher posting leans the other way, asking mainly for strong Python and mathematical thinking, with data science or ML experience typical but not required.
Developers are screened like elite software engineers, with a systems tilt. Hudson River Trading describes its process as a take-home test, about two phone interviews, and a full onsite day, assessing whether you write idiomatic, resource-efficient code in C++ or Python and can explain your reasoning as you go. TechPrep's 2026 write-up of the loop describes four to six back-to-back onsite rounds covering competitive programming, low-level systems, CS fundamentals, and behavioural questions, and stresses that flawless implementation matters as much as the idea. Know-Quant sums up the contrast: developer interviews ask precise questions with a right answer; researcher interviews ask questions without one.
Switching seats later
People plan to "start as a dev and move to research." It happens, but do not bank on it. In an April 2023 Blind thread, a Citadel quant developer on a reported $400K package asked about moving into research; Citadel colleagues replied that the firm is merit-based and the move works when a seat is open, the skills match, and you are a top performer in your current role. In an older Blind thread on the same question, a Two Sigma employee called it "possible but not common." Know-Quant describes the usual path as demonstrating research work informally before asking for the seat, and says the reverse move, researcher to developer, is rarer because production engineering takes years to learn.
Trader-to-researcher moves follow similar logic: you need a skill the target team cannot easily hire. If your goal is research, apply to research. Treat the developer route as a fallback that pays well, not a side door.
Which should you choose?
- You are fast at mental math, like games, and can stomach a volatile bonus: trader. The steepest curve and the most undergrad-friendly entry, but the interview is unforgiving and the exit market is narrow.
- You have, or are finishing, a PhD with real research output: researcher. Your degree is worth $30K-$80K at entry (Quantt), and it is the ticket at funds like Citadel that require it.
- You are an undergrad who wants research without a PhD: target firms that do not gate on the degree. Jane Street calls a PhD "a plus" for its New York researcher role, and D.E. Shaw hires bachelor's and master's quant analysts on its own pay line.
- You are a strong C++ or systems engineer: developer. You keep big-tech-portable skills and still earn $300K-$500K in year one at the top market makers (Quantt's HRT estimate).
- You are an international student who needs a visa: weight portability. A developer skill set finds a new US or UK employer quickly if a visa lottery or a desk closure goes against you; a trader's skills mostly transfer to other trading firms. That is worth more than the extra bonus in your first two years.
Practice the loop you will actually face
The three exams barely overlap, so prep for the seat, not for "quant." For trading, drill the speed screen with the Optiver trader mock interview or the SIG quantitative trader loop. For research, run the Jane Street quantitative researcher mock, which follows that firm's probability-heavy format. For developer roles, the Citadel software engineer mock covers the systems-leaning coding rounds.
If you are torn between a quant offer and big tech, our Jane Street vs Google L4 breakdown runs the same pay-versus-portability math. And because researcher and trader applications open on different clocks, track which 2027 roles are open on the New Grad 2027 tracker.
Sources
- Quant Blueprint, "Quant Salaries: Published Base Pay by Firm & Role", postings checked Sept 28, 2026
- Quant Blueprint, Jane Street published base pay, checked Sept 28, 2026
- Jane Street, Quantitative Researcher (New York) job posting, retrieved Oct 2026
- Harvard FAS Career Services, Citadel Quantitative Researcher, PhD Graduate (US) listing, recruitment opened July 13, 2026
- Quantt, "Quant Salary 2026: How Much Do Quants Make in the US?", updated Sept 24, 2026
- Quantt, "Quant Researcher Salary 2026", updated Aug 6, 2026
- Quant Blueprint, "Quant Researcher vs. Trader vs. Developer", retrieved Oct 2026
- Know-Quant, "Quant developer vs quant researcher: which seat suits you", Jan 2026
- EverythingQuant, "Optiver Trader Interview Guide", retrieved Oct 2026
- Hudson River Trading, "How to Prepare for Your Software Engineer Interview at HRT", Sept 2021
- TechPrep, "Hudson River Trading's Interview Process (2026)"
- Blind, "Citadel HFAM: quant dev to quant research", Apr 2023
- Blind, "Chance from quant developer to quant researcher", Jul 2019
Frequently asked questions
- Which pays more: quant trader, quant researcher, or quant developer?
- At entry, base pay barely separates them: Jane Street's own New York postings list a $300,000 base for its quantitative trader, quantitative researcher, and software engineer roles alike (Quant Blueprint, checked Sept 28, 2026). The gap opens in the bonus and the ceiling. Quantt's US guide (updated Sept 24, 2026) puts senior traders and researchers at $700K-$2M+ total and senior quant developers at $500K-$900K, with portfolio managers at $1M-$10M+.
- Do you need a PhD to be a quant researcher?
- At many hedge funds, yes. Citadel's 2027 Quantitative Researcher posting, which opened July 13, 2026, requires a PhD in a quantitative field and independent research experience. Jane Street's New York researcher posting calls a PhD or research experience 'a plus' rather than a requirement. Quantt's researcher guide (updated Aug 2026) says junior PhD researchers start $30K-$80K higher in total comp than master's hires.
- Is quant trading easier to get into than quant research for undergrads?
- It is the more natural entry point, not an easier one. Trader roles are typically filled by bachelor's and master's graduates (Quant Blueprint), but the screen is a speed gate: Optiver's online assessment includes 80 arithmetic questions in 8 minutes, a 26-question sequence test in 25 minutes, and a 10-question probability test (EverythingQuant). Researcher roles at PhD-gated funds are closed to most undergrads regardless of ability.
- Can a quant developer move into quant research?
- Yes, but it is the exception. In a 2023 Blind thread, Citadel employees said a developer gets a shot at research if a seat is open, the skills match, and they are a top performer; in an older Blind thread, a Two Sigma employee called the move 'possible but not common.' Know-Quant notes the developer-to-researcher move usually follows informally demonstrated research work, and the reverse move is rarer.
- Which quant role is safest for an international student on a visa?
- Quant developer, on portability. A C++ or Python systems skill set transfers to big tech and thousands of other employers if a US visa lottery or a desk closure goes against you. Trader and researcher skills are more firm-specific, and trader pay is the most volatile. That safety costs upside: Quantt places senior developer comp at $500K-$900K versus $700K-$2M+ for senior traders and researchers.