Your Susquehanna Prep Plan

Susquehanna runs two different hiring pipelines under one name, and the first job of any prep plan is to make sure you are on the right one. The famous SIG loop is the quantitative-trader track: a probability and expected-value math test, make-a-market games with cards and dice, and mental math at speed. It is not a coding interview at all. This plan is for the other track, the software-engineer and quant-developer loop. That one is a genuine coding pipeline: a CodeSignal General Coding Assessment, then one or two live technical rounds, then a Super Day. Interviewing to trade? Close this and go drill probability and market-making instead. If you are interviewing to build, the good news is that SIG reskins a small, recurring set of problems rather than a firehose of random algorithms. A focused plan goes a long way.

The Susquehanna angle

Two facts shape everything downstream. First, the SWE assessment is a standard CodeSignal General Coding Assessment, reported at roughly four questions in about seventy minutes, easy to medium. Intern and older reqs have run shorter, so confirm the exact shape with your recruiter rather than assuming. Second, SIG rewards clean data-structure choice and correct stateful implementation over clever optimums. The single most-reported technical problem is a range-and-category lookup store, the "movie database." Candidates over-reach for segment trees there; the interviewer just wants hashmap indexing. Read the interview process and how hiring decisions get made first, so the two-track split and the Super Day are clear before you plan around them.

The plan, phase by phase

Each phase below carries one focus and a checkpoint that tells you when you have earned the next. A phase is a share of your total prep time, not a fixed number of weeks, so the plan holds whether your runway is two months or one hard week. The proportions are the message. The CodeSignal assessment leads, because it is the pass-or-fail gate and nothing downstream counts until you are through it, and its pass rate is widely understood to be low. The recurring problem set does double duty, since the same archetypes that clear the assessment come back in the live round, so time spent there compounds. Design sits light on purpose, because for most SWE reqs SIG "design" means OOP and in-memory optimization, not distributed-systems whiteboarding. Do not step past a checkpoint until you can clear it cold.

To make the proportions concrete, picture a four-week runway. The first days go to Phase 1: confirm you are on the coding track and map the loop with your recruiter. Week one and half of week two belong to the assessment. Build the range-and-category store, the lazy-update hashmap, the ticker trie, and the simulation and OOP problems until each runs on reflex. These four are most of what CodeSignal and the live round throw at you. The back half of week two is the technical round, working the CSV-optimize problem and narrating complexity until refactoring naive code is automatic. Week three is the light design section, the in-memory market-data store carried end to end, scaled to your level. Week four is fit stories, a true "why SIG," a mock loop or two, and skimming the offer playbook so nothing catches you flat. On a tighter runway you compress the same order without reordering it, because the assessment comes before everything.

Why the phases are weighted this way

The weighting tracks how SIG filters. The assessment leads. It is a hard pass-or-fail gate with a low pass rate, and nothing downstream counts until you are through it. The recurring problem set takes the largest combined share because it is unusually concentrated: the same four archetypes clear the assessment and then reappear in the live round, so hours spent there pay twice. The technical round takes a real band. Its optimize-the-naive-code framing is a specific skill that blank-whiteboard practice leaves you least ready for. Design sits light because for most SWE reqs it is OOP and in-memory work rather than distributed systems, and pushing it hard would steal time from the coding that decides the loop. Fit and offer close the plan. The behavioral is a genuine conversation, not a formality, and the no-equity comp shape rewards knowing the levers before you are asked for a number.

The Susquehanna angle

The trap is preparing for SIG as if it were either the trader track or a generic FAANG loop. People grind probability and market-making for a coding req, or grind hard LeetCode for an assessment that rewards picking a hashmap over a segment tree and implementing a stateful system correctly. Or they whiteboard distributed systems for a round that is really about indexing an in-memory store. Spend your hours where SIG looks. Get the range-and-category store and the other three archetypes into your hands until they flow, make optimizing naive code automatic, keep design light unless your level demands otherwise, and bring honest, specific fit stories. Those pieces turn SIG from a curveball into home ground.

In the room

• Confirm you are on the coding track first. The trader loop is math, expected value, and market-making games with no coding; poker is post-hire culture, not a round. This plan is for the SWE / quant-developer CodeSignal pipeline.
• Clear the CodeSignal assessment before anything else. It is a standard General Coding Assessment, reported around four questions in seventy minutes, easy to medium. The bar is clean, correct data-structure choice under a clock, not hard dynamic programming.
• Drill the four recurring archetypes until they are automatic. The range-and-category lookup store is the signature problem; the lazy-update hashmap, the ticker trie, and the simulation-and-OOP builds round out the set, and they return in the live round.
• In the technical round, optimize the naive code and speak the complexity. Expect to read an existing linear scan, often a CSV movie search, and refactor it to an indexed lookup in a shared editor. Think out loud and test before you finish.
• Bring honest fit stories and know the offer shape. The behavioral is a resume-and-fit conversation grounded in unnumbered values; the offer is a high base plus cash bonus plus sign-on with essentially no RSUs, so the cash levers are what you negotiate.