Jane Street Software Engineer Interview Guide

The short answer: Jane Street's software engineer interview is coding from start to finish. Its prep page and 2020 blog post describe a remote technical interview, then a final round of 2 to 4 coding interviews, now held in person at an office. Jane Street says the rounds include no mental math, logic puzzles or probability questions, and writing OCaml earns no bonus. Each round is one open-ended problem that grows in parts, and interviewers care more about how you work through it than whether you finish. Last reviewed October 2026.

Jane Street software engineer interview: quick answers

The Jane Street software engineer role

What Jane Street engineers work on, the skills its job postings ask for, and what a normal week looks like.

Jane Street software engineer levels and pay

Your level sets the bar you are interviewed against and the pay band of the offer.

Jane Street does not publish an engineering ladder. It posts one Software Engineer title for new grads and experienced hires at the same New York base, and staff describe a flat hierarchy. Candidate guides use level labels such as L1, which Jane Street itself does not use, so the table below uses posting titles.

Pay in Jane Street's US job postings

Most of Jane Street's New York engineering postings print a single fixed base of $300,000, not a range. Each says total pay also includes an annual discretionary bonus. No equity is mentioned.

The new grad and experienced Software Engineer postings print the same base, and none of the postings state a level. Production, Linux, data and hardware engineering postings are left out because they sit outside the general software engineer track. Ranges seen October 27, 2026.

Jane Street software engineer interview process

Jane Street's prep page, interviewing page and blog posts from 2014, 2017 and 2020 describe the coding interviews. The recruiter call, round lengths and the project discussion come from candidate reports.

Based on Jane Street's own pages and candidate reports.

Jane Street aims to reply within about a week of the final interviews and says it does not make exploding offers, so flag any competing deadline early. Candidate reports describe team matching after the on-site. Last reviewed October 2026.

What is different at Jane Street

• Coding only, with no math. Jane Street says software engineer interviews skip the mental math, logic puzzles and probability questions used in its trading interviews.
• One problem that keeps growing. The 2017 blog post gives an example: memoize a function, then bound the cache, then add LRU eviction. Jane Street says it avoids trick questions that need a single insight.
• OCaml earns nothing extra. Jane Street runs almost everything on OCaml, yet its prep page says there is no bonus for functional languages and to avoid OCaml unless it is your strongest language.
• Finishing is optional. The 2014 and 2017 blog posts say clear communication and careful, correct code count for more than reaching the last part, and a careful candidate who gets less far can beat a sloppy one who finishes.
• Jane Street publishes its own practice material: a 34-minute mock interview video with two engineers working a retired question, and a worked question in the 2017 blog post.
• Fixed pay in postings. Most New York software engineer postings print a single $300,000 base, the same for new grads and experienced hires, plus a discretionary bonus.

Jane Street software engineer interview questions

Coding problems from our bank tagged to Jane Street, then design, fundamentals and behavioral prompts we wrote for this guide.

1from typing import Dict, List, Tuple
2
3def track(fills: List[Tuple[str, int, int]]) -> Dict[str, Tuple[int, int]]:
4    """Return {symbol: (net position, realized profit in cents)} using FIFO lot matching."""
5    pass

1from collections import defaultdict, deque
2from typing import Dict, List, Tuple
3
4def track(fills: List[Tuple[str, int, int]]) -> Dict[str, Tuple[int, int]]:
5    lots = defaultdict(deque)  # each lot is [signed qty, price]; all open lots share one sign
6    realized = defaultdict(int)
7    for symbol, qty, price in fills:
8        book = lots[symbol]
9        realized[symbol] += 0
10        while qty and book and (book[0][0] > 0) != (qty > 0):
11            lot = book[0]
12            matched = min(abs(qty), abs(lot[0]))
13            if lot[0] > 0:  # a sell closes a long lot
14                realized[symbol] += matched * (price - lot[1])
15                lot[0] -= matched
16                qty += matched
17            else:  # a buy closes a short lot
18                realized[symbol] += matched * (lot[1] - price)
19                lot[0] += matched
20                qty -= matched
21            if lot[0] == 0:
22                book.popleft()
23        if qty:
24            book.append([qty, price])
25    return {s: (sum(lot[0] for lot in lots[s]), realized[s]) for s in realized}
26
27# Complexity: O(n + k) time for n fills and k lots opened, since each lot is closed at most once;
28# O(k) space for open lots.

How to prepare

A plan for the Jane Street loop, most important first:

1. Practice open-ended problems that grow in parts: solve a first version, then add a new requirement and extend your code without starting over.
2. Pick the language you know best and drill its standard library and basic data structures, since Jane Street asks for real code and gives no credit for OCaml.
3. Watch Jane Street's mock interview video and work the retired question from its 2017 blog post under a timer.
4. Practice talking through every step aloud, agreeing on a plan before you type and saying clearly when you do not know something.
5. Check your code by hand on small inputs and edge cases, since careful, correct code beats getting further with bugs.
6. If you have several years of experience, prepare one project you can discuss in depth, and ask your recruiter whether AI tools are allowed in any round.

Practice path

Courses that cover the rounds above:

• Data Structures and Algorithms
• Python Basics