D.E. Shaw MTS India January 2025: Quant MCQs, DSA, SQL, and the Batch Hiring Machine

TL;DR. 2023 graduate, selected for D.E. Shaw's Member Technical (MTS) position in India, January 2025. The online assessment was unlike anything at a standard tech company: 3 medium DSA coding problems alongside 10 math/probability MCQs and 10 CS fundamentals MCQs. The math section tested combinatorics, probability distributions, and expected value calculations. Technical interview rounds covered SQL (fetching first login time per userLoginId from a history table), C++ macros, method hiding in inheritance hierarchies, and a low-level design discussion modeled on Stack Overflow. D.E. Shaw interviews in batches and shortlists from the pool, so the timeline is slow and the candidate pool is curated. Source post linked at the bottom.

This is not a "D.E. Shaw placement preparation" overview. Those list the eligibility criteria and stop. This is one specific January 2025 candidate's specific loop, with the actual DSA problems, the actual math question types, and the actual technical topics. The distinctive feature: the quant MCQ section that no FAANG, no Indian startup, and no service company includes.

Why D.E. Shaw's hiring process is different from every other tech company

D.E. Shaw is not a technology company. It is a quantitative investment firm that employs technology to trade financial markets. This distinction shapes every aspect of their hiring process.

At Google, Amazon, Flipkart, or Swiggy, the interview evaluates: can this person write correct code, design systems, and work with a team? At D.E. Shaw, the interview evaluates all of that, plus: does this person have the mathematical intuition to work in an environment where software decisions have direct financial consequences?

The result is an online assessment that includes a section you will not find anywhere else: math and probability MCQs. Not "aptitude" in the TCS/Infosys sense (percentages, time-and-work, pipes-and-cisterns). Actual probability theory, combinatorics, and expected value problems that require formal mathematical reasoning.

If you are preparing for D.E. Shaw and you skip the math section because "I am applying for a software role, not a quant role," you will fail the online assessment. The math MCQs are a filter, and they carry enough weight to disqualify otherwise strong DSA candidates.

The MTS candidate profile

D.E. Shaw's MTS role is their entry-level engineering position for candidates with 0 to 2 years of experience. It is distinct from their quantitative analyst and quantitative developer roles, which require stronger mathematical backgrounds. MTS engineers build and maintain the firm's trading infrastructure, risk systems, and internal tools.

The 2023 graduation year means this candidate had approximately 1.5 years of post-graduation experience at the time of the January 2025 interview. D.E. Shaw does not require prior experience for MTS: fresh graduates from top engineering colleges are common in their hiring pool. But the mix of 1.5 years of industry experience plus DSA and math strength made this candidate competitive.

The online assessment: 3 sections, 1 sitting

DSA problems

Two of the three DSA problems were identified in the source:

Problem 1: Lexicographically smallest string after k deletions

Given a string and an integer k, return the lexicographically smallest string possible after deleting at most k characters.

Technique: Stack-based greedy approach. Iterate through the string. For each character, while the stack is non-empty and the top of the stack is greater than the current character and k > 0, pop the stack and decrement k. Push the current character. After iteration, pop remaining k characters from the end. This is a variant of LeetCode 402 (Remove K Digits).

Time complexity: O(n) where n is string length.

Problem 2: Count inversions with 2x condition

Given an array, count pairs (i, j) where i < j and arr[i] >= 2 * arr[j].

Technique: Modified merge sort. During the merge step, count pairs where the left element is at least twice the right element, then merge normally. This is a variant of LeetCode 493 (Reverse Pairs).

Time complexity: O(n log n).

Both problems are in the medium-to-hard range and require familiarity with specific algorithmic patterns (monotonic stack, modified merge sort). A candidate who has only practiced array/string manipulation and basic tree problems will struggle here.

The third DSA problem was not specified in the source post.

Math and probability MCQs

The exact questions were not reproduced (NDA or memory constraints), but the category is well-documented across D.E. Shaw interview reports:

These are not JEE-level aptitude questions. They require formal probability knowledge at the level of a first course in probability theory (Sheldon Ross or equivalent). If your last encounter with probability was in Class 12 NCERT, this section will disqualify you.

Preparation recommendation: Work through the first 6 chapters of "A First Course in Probability" by Sheldon Ross. Solve 50+ probability problems from competitive math sources (Brilliant.org, AoPS, or probability sections of quantitative aptitude books like Brainstellar). Practice under timed conditions: 10 problems in 15 minutes.

CS fundamentals MCQs

10 questions covering: Operating Systems (process scheduling, memory management, deadlocks), DBMS (normalization, transactions, indexing, SQL concepts), Computer Networks (TCP/IP, HTTP, DNS, routing), and OOP (polymorphism, inheritance, encapsulation, design patterns).

This section is similar to what Goldman Sachs and Morgan Stanley include in their online assessments. The questions are conceptual, not implementation-based. If you have studied CS fundamentals for any campus placement drive, this section is familiar territory.

Technical interview rounds

D.E. Shaw's interview rounds after the online assessment cover multiple technical axes. The candidate described the following topics across the rounds (individual round boundaries were not clearly delineated in the source):

SQL: First login time per user

Problem: Given a table with columns (userLoginId, loginTime), fetch the first login time for each unique userLoginId.

Solution:

SELECT userLoginId, MIN(loginTime) AS firstLogin
FROM login_history
GROUP BY userLoginId;

This is a straightforward GROUP BY with an aggregate function. The interviewer likely followed up with: what if you also need the session details from that first login? That requires a self-join or window function:

SELECT *
FROM login_history h
WHERE loginTime = (
    SELECT MIN(loginTime)
    FROM login_history
    WHERE userLoginId = h.userLoginId
);

Or using ROW_NUMBER:

SELECT *
FROM (
    SELECT *, ROW_NUMBER() OVER (PARTITION BY userLoginId ORDER BY loginTime) as rn
    FROM login_history
) t
WHERE rn = 1;

D.E. Shaw tests SQL because their infrastructure runs on relational databases. Trading systems, risk systems, and compliance reporting all rely on complex SQL queries. An MTS engineer who cannot write correct SQL will struggle from week 1.

C++ macros

Macro expansion is a C-preprocessor topic that D.E. Shaw tests because their legacy codebases (some dating to the 1990s) use macros extensively. The interviewer likely asked about: macro expansion order, the difference between #define and const, why macros can produce unexpected results (missing parentheses in expressions, multiple evaluation of arguments), and the ## token-pasting operator.

This is niche knowledge that most modern developers do not use. If you are interviewing at D.E. Shaw and your primary language is Java or Python, you still need to understand C/C++ preprocessor behavior. It appears in their interviews consistently.

Method hiding in inheritance

Method hiding (as distinct from method overriding) occurs when a derived class defines a static method with the same signature as a static method in the base class. In C++, this also applies to non-virtual functions: if the base class function is not declared virtual, the derived class function hides it rather than overriding it.

The interviewer likely tested: the difference between hiding and overriding, what happens when you call a hidden method through a base class pointer, and the role of virtual in enabling polymorphic dispatch.

Low-level design: Stack Overflow

Problem: Design the core entities and interactions of a Stack Overflow-like Q&A platform.

Expected scope: Users, Questions, Answers, Comments, Votes (up/down), Tags, Reputation system, Search. The LLD should cover: class hierarchy, relationships (a Question has many Answers, an Answer has many Comments, a User has Reputation), voting mechanics (how reputation changes on upvote/downvote/accept), and the moderation system (flagging, closing, duplicate detection).

This is a mid-complexity LLD problem. The interviewer is evaluating: can you identify the right entities, define clean interfaces between them, and handle the reputation system's business logic (which has non-obvious rules like: accepting an answer gives +15 to the answerer and +2 to the asker).

The batch hiring machine

D.E. Shaw does not hire continuously. They hire in batches. The process works like this:

1. Applications collect over a period (typically 2 to 4 months).
2. Online assessments are scheduled for the batch. Everyone in the batch takes the same assessment within a narrow window.
3. Top performers from the assessment are shortlisted for interviews.
4. Interviews happen over 2 to 4 weeks, conducted by D.E. Shaw engineers in Hyderabad (their primary India office).
5. Offers are extended after the batch is fully evaluated.

This means the timeline from application to offer is slow: typically 3 to 6 months. Candidates who apply and expect a response within 2 weeks (as they would from Amazon or Google) become frustrated. The slow pace is structural, not a signal of disinterest.

The batch model also means that D.E. Shaw compares candidates against each other within the batch, not against an absolute bar. Your assessment score is ranked against everyone else in the same batch. A score that would clear the bar in a weak batch might not clear it in a strong batch.

Practical implication: You cannot time your D.E. Shaw application to coincide with a weak batch. Apply when you are ready, not when you think the competition will be soft.

How D.E. Shaw compares to Goldman Sachs for engineering roles

Both are financial institutions that hire engineers. The comparison is common among candidates:

If you are considering both, the key differentiator is the math requirement. Goldman Sachs engineering roles can be cleared with strong DSA and CS fundamentals. D.E. Shaw MTS requires those plus genuine mathematical problem-solving ability.

The actual questions, compiled

How to actually prepare for D.E. Shaw MTS

1. Math is not optional. Sheldon Ross chapters 1 through 6. Brainstellar probability problems. Expected value and linearity of expectation must be second nature.

2. DSA at medium-hard level. LeetCode 402 (Remove K Digits), 493 (Reverse Pairs), and their variants. D.E. Shaw's DSA problems use specific patterns (monotonic stack, modified merge sort, segment trees) more than brute-force or simple DP.

3. SQL is tested. Write queries from scratch, not with ORM help. Window functions, self-joins, GROUP BY with HAVING. Practice on HackerRank SQL section.

4. C++ internals if you list C++. Macros, virtual dispatch, multiple inheritance, smart pointers, move semantics. D.E. Shaw's codebase is C++-heavy.

5. Expect a slow timeline. 3 to 6 months from application to offer. Do not put other interviews on hold while waiting.

How we checked

PapersAdda's verification standard requires a publicly accessible post URL, per-round detail from the candidate, and a stated outcome. The LeetCode Discuss post meets these criteria: it specifies the role (MTS), the location (India), the month (January 2025), the graduation year (2023), the OA structure (3 DSA + 10 math + 10 CS MCQs), the specific DSA problems, the interview topics (SQL, C++ macros, method hiding, Stack Overflow LLD), and the outcome (offer).

Source: D.E. Shaw Interview Experience, MTS, India, January 2025, Offer on LeetCode Discuss.

Limitation: LeetCode Discuss posts are sometimes edited or removed. The data was captured in May 2026. If the original post is no longer accessible, the details above reflect its state at time of capture.

Related on PapersAdda

• Goldman Sachs Engineering Analyst 2025: 10.5-Hour Campus Marathon, the investment-bank counterpart to D.E. Shaw's hedge-fund hiring
• Google L4 India: 7-Round Downlevel to Rejection, when process length does not correlate with outcome
• Uber SDE-1 India 2025: Google Sheets LLD Rejection, LLD communication skills deciding the outcome
• Swiggy SDE-2 Backend 2025: ₹40 LPA, Rejected at Train Booking LLD, hybrid rounds that test DSA and design simultaneously

Related: Infosys SP/DSE coding questions 2026, for the full topic-frequency breakdown and 47 practice questions with solutions.

Related: LeetCode questions asked in TCS 2026, for the verified question-frequency analysis and pattern-wise prep approach.

A note on sourcing: details on this page are candidate-reported and assembled from public preparation resources, so exact numbers shift between drives. Always verify the live notification on the official site before you apply or make a decision.