Is Jump Trading or HRT harder to get into?

Both hire around 100 graduates a year out of tens of thousands of applications, per Quantt's 2026 firm guides. The screens differ: Quantt puts HRT's first-round pass rate at roughly 5-10%, Jump's at 10-15%, and Glassdoor candidates rate HRT's process harder (3.45/5 vs Jump's 3.19/5, retrieved September 2026). If you code fast and prove things cleanly, HRT is the tighter gate.

Which pays more, Jump Trading or HRT?

levels.fyi shows the same $500K median software-engineer package at both firms (retrieved 6 September 2026), so the headline gap is close to zero. The structure differs: Jump's median splits $250K base and $250K bonus with no equity, while HRT's reported range runs from $446K at L1 to $723K at L3. Firm-wide performance matters more than the level ladder at either shop.

Does HRT's record revenue mean more new-grad hiring?

Not directly. HRT posted $6.4B of Q1 2026 trading revenue and $11.4B in Q2 (Hedgeweek, May 2026; Cryptobriefing, August 2026), but Revelio Labs put its March 2026 headcount at about 1,490 against Jump's 2,137. HRT converts good years into trading capital and bonuses rather than headcount, so seats stay scarce.

Which firm is better for an international student?

Jump, on optionality. Jump lists 13 offices across 8 countries and is actively building a statistical-arbitrage team in Hong Kong (Cryptobriefing, September 2026), so a non-US seat is a real path. HRT lists 14 offices but concentrates 64% of staff in North America (Revelio Labs, March 2026), which in practice means the H-1B lottery.

Do I need C++ for both?

For engineering roles, yes. Both firms run C++ as the production language, and both onsites include a dedicated low-latency systems round covering lock-free structures, memory ordering and cache behaviour (techinterview guides, 2026). Research roles are more forgiving — Python plus statistics carries you further at Jump's quant-researcher loop.

Jump Trading vs HRT (2026): Interviews, Comp, Culture

Jump Trading and Hudson River Trading are the two firms new grads most often hold offers from at the same time, and the comparison that gets made — "which one pays more" — is the least useful one. On the public numbers they pay about the same: levels.fyi lists a $500K median software-engineer package at both (retrieved 6 September 2026). The real fork is structural. HRT is a smaller, New York-weighted firm having an extraordinary year; Jump is a larger, more geographically spread firm building new desks in Asia.

That difference shows up in how many people they hire, where the seats are, and what the onsite actually tests.

The 30-second verdict

Jump Trading	Hudson River Trading
Headquarters	Chicago	New York
Headcount (Mar 2026, Revelio Labs)	~2,137	~1,490
Offices	13 across 8 countries (Cryptobriefing, Sep 2026)	14 worldwide (HRT careers page, retrieved Sep 2026)
North America share of staff	50.5%	64.0%
Recent business news	Hong Kong stat-arb build-out; 99,305 sq ft NY lease, May 2026	$6.4B Q1 2026 revenue, $11.4B Q2 2026
First-round pass rate (Quantt, 2026)	~10-15%	~5-10%
Onsite	5-7 rounds, Chicago / NY / London	5-6 rounds in one day, usually NY
Grad Y1 package (Quantt, 2026)	$250K-$500K	$300K-$500K
SWE median TC (levels.fyi, Sep 2026)	$500K	$500K
Glassdoor interview difficulty	3.19 / 5	3.45 / 5
Positive interview experience	60.3%	46.5%

The year each firm just had

HRT's 2026 is the more dramatic story. The firm booked about $6.4B of trading revenue in Q1 — roughly 135% up year on year, with around $4.2B of profit and roughly $20B of net trading capital deployed (Hedgeweek, May 2026). Q2 was larger still at $11.4B, against a full-year 2025 figure of about $12.3B (Cryptobriefing, August 2026). In other words, HRT earned close to a whole 2025 inside three months of 2026.

Jump's news is about surface area rather than a single number. It signed a 99,305 sq ft New York lease in May 2026 and is expanding its Hong Kong office, where it is building out a statistical-arbitrage team focused on low- and mid-frequency strategies, with recruiting led by a former Citadel Securities hire (Cryptobriefing, September 2026). Jump also runs the wider student funnel: Extern's guide counted 11 US intern requisitions live on 15 August 2026 across eight tracks, including three AI research-engineer variants, FPGA and systems engineering.

The practical read for a candidate: a record quarter changes your bonus, not your odds of getting hired. Headcount is the better proxy for seats, and Jump carries about 640 more people.

Interviews: same bar, different shape

Both loops start with a timed coding assessment and end with a team match after the offer, so the offer is for the firm, not a desk. Between those points they diverge.

HRT is a depth-of-fundamentals loop. Quantt describes an assessment of two or three hard problems in 90 minutes, one 60-minute CoderPad screen, then five to six interviews in a single day, usually in New York, with offers landing within about two weeks. techinterview's 2026 guide breaks the onsite into two or three algorithm rounds (hard graph problems, state-compressed DP, computational geometry with floating-point edge cases), a dedicated low-latency systems round covering kernel bypass, lock-free structures, memory ordering and NUMA, and a probability round of expected-value and Bayesian problems. It puts the competitive baseline at USACO Gold or Codeforces 1800+.

Jump is a wider funnel with a longer day. Quantt puts the online assessment at 90-120 minutes, then two or three technical phone screens of 45-60 minutes with working professionals rather than recruiters, then a Super Day of five to seven back-to-back interviews in Chicago, New York or London, with the hardest architectural and research-design rounds deliberately scheduled last. techinterview's Jump guide adds a trading-specific round for trader candidates — CME futures microstructure, contract specifications, options pricing, and crypto mechanics for Jump Crypto roles — plus a manager or director round that HRT's loop does not mirror.

The asymmetry candidates report is in tone, not difficulty. On Glassdoor, 60.3% rated Jump's process positive against 46.5% for HRT (retrieved September 2026), while Jump's average process runs about 24 days across 187 submitted interviews. Both guides describe C++ as effectively required for engineering roles; Rust turns up in Jump's newer systems.

Money, and the part the headline number hides

levels.fyi's September 2026 pages put both firms at a $500K median for software engineers. Jump's median package splits $250K base and $250K bonus with no equity component — private firm, no stock. HRT's page shows a range from $446K at L1 to $495K at L2 and $723K at L3.

Read that against what candidates say, because the two disagree. Quantt puts HRT graduate engineers at $300K-$500K in year one and Jump graduates at $250K-$500K — both below the levels.fyi L1 figure. That gap is the usual self-selection in voluntary salary reporting, and it is worth budgeting around.

The other thing worth reading closely is payout timing. In a Blind thread from November 2022, a poster estimated HRT algo-engineering packages at $400K-450K and described bonus and signing money paid quarterly over about 18 months; an HRT employee replied in the same thread that the deferral structure as described was not accurate and told the candidate to ask the recruiter. Both halves of that exchange are the lesson: deferral schedules at these firms are real, they vary, and they are a recruiter question rather than a forum question. A 2025 Blind thread comparing the two firms landed in the same place — one experienced trader wrote that different teams are almost like different firms, which makes team match, not the offer letter, the decision that sets your first two years.

Culture, hours and where you would actually sit

Posters in that 2025 thread describe both firms as demanding in the same way: 50-60 hour weeks, small teams, individual contribution driving the bonus, and progression that depends far more on your desk than on the logo. Nobody in the thread argued that either firm has better work-life balance than the other.

Geography is the cleaner differentiator. HRT concentrates 64% of its people in North America, with Northern Europe at 14.5% and East Asia at 8.1% (Revelio Labs, March 2026), and its careers page lists 14 offices from NYC and Austin to Singapore, Shanghai, Mumbai and Dublin. Jump splits closer to half and half: 50.5% North America, 17.9% Northern Europe, 8.5% East Asia, out of about 2,137 people. If you want a real chance at a London, Singapore or Hong Kong seat, Jump's distribution and its current Asia hiring make that a more plausible ask.

Which should you choose?

You are an international student who would rather not sit in the H-1B lottery. Jump. The Asia and Europe headcount is proportionally larger and the Hong Kong desk is actively hiring quantitative researchers (Cryptobriefing, September 2026).
You want the deepest low-latency engineering work and you are strong at competitive programming. HRT. Its onsite is built around exactly that, and USACO Gold / Codeforces 1800+ is a fair self-test before you apply.
You are a trader or a researcher rather than an engineer. Jump. Its loop has a dedicated trading round and eight separate student tracks, including PhD-specific research and AI research-engineer paths (Extern, August 2026).
You have both offers and similar teams. Ignore the headline TC — they are within noise of each other — and interrogate the deferral schedule and the desk's P&L instead.
You have one offer and one pending onsite. Take the timing seriously: HRT returns decisions in about two weeks, Jump in roughly 24 days on average (Quantt 2026; Glassdoor, retrieved September 2026).

Prepare for both

The overlap between these loops is large enough that one preparation block covers both: hard algorithms under time pressure, C++ internals, one probability round, one low-latency systems conversation.

Run the researcher loop end to end with Applr's HRT quantitative researcher mock, which follows the live CoderPad format the phone screen uses.
If your Super Day includes a market-facing round, the Citadel quantitative researcher mock and the Jane Street quantitative researcher mock drill the same expected-value and market-making reasoning Jump's trading round uses.
Compare HRT against a trader-first shop in HRT vs Optiver (2026) before you decide the culture question.
Check your resume clears an automated screen with the free resume checker, then track deadlines on the New Grad 2027 tracker.

Jump Trading vs HRT (2026): Interviews, Comp, Culture

The 30-second verdict

The year each firm just had

Interviews: same bar, different shape

Money, and the part the headline number hides

Culture, hours and where you would actually sit

Which should you choose?

Prepare for both

Sources

Frequently asked questions

Ready to land more interviews?

Practice for top companies

More from the blog