Introduction

I’m thrilled to share my experiences during the 2024 Internship season and the valuable lessons I’ve learned along the way. I’ll be discussing my preparation, experiences, and advice for both Quant Roles and SWE roles at different companies.

Quant Roles will be my main focus since they were my top priority, but I’ll also share how I prepared for SWE roles, as the tests for both overlap quite a bit. We’ll break down our discussion into three main parts:

1. Preparation
2. Tests
3. Interviews

Let’s get started!

Preparation

Quant Roles

The following are the resources which I completed for Quant Roles:

1. BrainStellar:

   • These questions are really good! Start with the easy section. Many companies ask the exact same questions from this resource.

2. Puzzled Quant:

   • Questions are a bit vague and sometimes incomplete.
   • Solutions were absent for a few problems, but the ideas used in these problems are a superset of almost all techniques covered in standard quant preparation books.
   • Also has “80 in 8” and “Find Next in Sequence” games (only Optiver asked them in my year).
   • For mental maths, use Zetamac or Trading Interview.

3. Quant Guide:

   • An alternative to the above. Found out about it a bit late, but peers have given excellent reviews (LeetCode for quant).

4. Xinfeng Zhou:

   • Best resource. Covers every topic which is generally asked.
   • Except for Chapter 6: Finance, all the chapters are highly relevant.
   • Only options pricing from Chapter 6 was asked by a company (Tower Research).
   • Can’t recommend this book enough!

5. Heard on The Street:

   • Also one of the best resources, and covers everything.
   • One of the top 2 books I would recommend.

I think these resources should cover all the questions you might have on the tests and in the interview. They’re pretty comprehensive, so you could almost take a whole summer to finish them (with SWE Preparation, of course). But make sure to at least get through the basics.

Here are the resources. I’d recommend them only to those who have already done the above quickly or have done them before while preparing for things like Jane SEE Program or Jane Street 2nd Year Internship test or whatever.

6. Additional Resources (if you’ve completed the above):
   • Mathematical Puzzles: Small number of problems, good for practicing at the end.
   • Fifty Challenging Problems in Probability: Good problems.
   • Martin Gardner: Contains some errors but decent.
   • Aniruddha Deb: The OG repo for CS students, covering almost every resource.
   • Quantitative Primer: Less known but good for post-standard books.
   • Dice Problems: Simply worded problems with excellent solutions.
   • Market Making: Very few companies like Optiver, Jane Street and Quantbox asked about market making. Companies usually provide some material for understanding about it.

SWE Roles

For this part of preparation, I mostly solved LeetCode problems and occasionally participated in CF contests. This is the easiest part, as most of the questions follow a very common pattern. We’ll discuss these patterns in the next section soon. By the way, many (most) of the quant roles also have a CP test in their first round and ask CP questions in interviews.

For SWE preparation, I mostly solved LeetCode problems and occasionally participated in CF contests.

1. CSES:

   • Highly recommended.
   • Great place to practice standard questions and learn well-known algorithms.
   • Focus on sections:
     • Introductory Problems
     • Sorting and Searching
     • Dynamic Programming (must)
     • Graph Algorithms (must)
     • Tree Algorithms (must)

2. LeetCode:

   • Good collection of problems, but don’t try to do everything. Follow a roadmap (e.g., NeetCode).

3. SDE Prep:

   • Good resource for CP practice.
   • I compiled the CP questions on this site during last semester for ANCC, DevClub, and MathSoc collaboration for Intern Prep Series.

In my opinion, these resources should be more than enough to prepare well for the tests. You can do a few practice tests on HackerRank or similar platforms that companies often use for their tests to get used to the platform. I personally find these platforms a bit confusing, though.

Key Topics

1. Dynamic Programming (DP)

• Most tests will have at least one question on DP, typically Knapsack or its variations.
• Practice recognizing patterns for DP problems. CSES very good. Leetcode Okayish. CF 1600-1700 rated good.

2. Binary Search

• Mostly covers the following:
  • Binary Search on sorted data.
  • Binary Search on the answer(Very frequent).
• Recognize Binary Search problems by checking:
  • Is the property monotonic?
  • Can you verify whether a value satisfies the property?
• Very good resource Erichto Video.

3. Tree/Graph Algorithms

• Easiest of all. All questions can be solved using these algorithms:
  • DFS/BFS
  • Spanning Tree
  • Dijkstra
• For tree-specific algorithms, practice Binary Lifting and Common Ancestor.
• CSES perfect for this. Leetcode is sometimes too simple. CF rated 1600-1700 ok.

Now we will be discussing the questions asked in the MCQ section of the test. This mostly covers OOPS. Questions usually involve being given a piece of code, and ask what will be the output for this. As a side note, remember that a few tests allow navigation between CP Problems and these MCQs. One may use this loophole to run the MCQ codes in the CP questions area, to see what the outcome is.

MCQ Section in Tests

Now we will be discussing the questions asked in the MCQ section of the test. This mostly covers OOPS. Questions usually involve being given a piece of code, and ask what will be the output for this. As a side note, remember that a few tests allow navigation between CP Problems and these MCQs. One may use this loophole to run the MCQ codes in the CP questions area, to see what the outcome is.

1. Inheritance Concepts

   • Most OOPs questions revolve around Inheritance, particularly in:

     • C++ (mostly).
     • Java.
     • Python (rarely).

   • Typical Question Format:

     • A parent class (P) and a child class (C) are given.
     • Both classes have:
       • A constructor with unique cout statements.
       • A destructor with unique cout statements.
       • A common method (e.g., foo) with unique cout statements.
     • The main function:
       • Initializes a child class pointer pointing to a parent class object.
       • Calls the foo method.
       • Ends with either:
         • Explicitly calling the destructor.
         • Allowing the default destructor to be invoked by the compiler.

   • Learn which class’s constructor, method, and destructor are invoked.

2. Buggy Code and Error Identification

   • Questions present a code snippet with logical or syntax errors.
   • Common errors:
     • Accessing or overwriting elements outside the bounds of an array.
     • Misuse of pointers or memory allocation issues.
     • Ambiguous method calls in multiple inheritance.

3. Structs and Padding

   • A struct with several attributes is provided, and the size of the struct is queried.
   • Key Points to Remember:
     • The size of a struct is usually the sum of the sizes of its attributes.
       • Example: If a struct contains 2 int variables, its size is 8 bytes.
     • When the struct includes data types such as strings or char, padding might be added to align memory.
       • Padding ensures attributes are stored in memory efficiently, based on the architecture’s alignment requirements.
     • Example:
       • For a struct with:

         struct Example {
             int a;      // 4 bytes
             char b;     // 1 byte
             float c;    // 4 bytes
         };

         • Due to alignment, padding is added after b to align c, making the total size 12 bytes instead of 9 bytes.
     • Tip: Read more about padding and alignment in the context of structs.

4. Additional Topics (Very few companies)

I mostly didn’t have much knowledge of these topics. So I guessed/left most of them. Although companies expect better from dual degree students. They may ask some senior for this.

• Computer Networks (COL334):
  • Simple questions about TCP/IP, OSI model layers, or protocols.
• Operating Systems (COL331):
  • Focus on deadlocks, scheduling algorithms, and basic memory management.
• Computer Architecture (COL216):
  • Questions on pipelines, caching mechanisms, and instruction cycles.

5. ML and Statistics

• Mostly covers statistics and ML questions.

• Common topics include:

  • Mean and manipulations of normal distribution (e.g., new standard deviation for sum of two independent variables).
  • Quartiles: Find 1st or 3rd quartile of given data.

• Asks questions like, which test is used when and so on. Like T-test, Chi-square test, etc.

• Learn basics of mean, quartiles, standard deviation, and hypothesis tests.

Tests

• Timing and Duration:

  • Tests often start right at the beginning of the semester.
  • Duration:
    • 1 hour for CP/SWE-focused tests.
    • 2 hours for tests covering both CP/SWE and MCQs.
  • Tests are usually scheduled between 6 p.m. and midnight.
  • This can be extremely taxing on both your mind and body.
  • The irregular schedule might lead to skipping meals, especially dinner, for several consecutive days.
  • It’s common to adjust your daily routine, such as I missed lectures for up to two weeks.

• Shortlisting and Frustrations:

  • Some companies delay announcing shortlists until the last moment, often 2-3 days before the placement/interview day.
  • This can be frustrating, especially after giving multiple tests without seeing immediate results.
  • Be prepared for surprises:
    • Example: A prominent quant company once shortlisted 14 candidates from the CS department but revoked the list 30 minutes before the deadline for ranking companies for interviews.
  • Some companies may prioritize diversity or other metrics over technical performance:
    • Candidates who did moderately well with low/average CG might get shortlisted over those who passed all test cases.
    • This behavior is because few companies know, that students with good CG and test performance, don’t prefer this companies.
    • Google, Amazon, Qube are few companies to follow this pattern. Bad luck if you consider them as your backup.

• Scholarship Programs

  • Some quant companies like Quadeye offer scholarship programs during the summer semester break.
  • The tests are relatively easy.
  • Interviews focus on:
    • Basic topics from COL106 (Data Structures and Algorithms).
    • Your projects listed on the CV.
    • Getting the scholarship often means an automatic shortlist for future interviews.

Interviews

• General Information:
  • Interviews typically kick off early in the morning on Day 0.
  • The first slot of interviews ends around noon.
  • You can select up to four companies for interviews during this slot.
  • If a company selects you during the interview, they generally inform you immediately if they’re extending an offer.
  • Sometimes some company don’t inform you on spot. This is generally the case when the company considers you as their backup.
    • Interviewers may ask about your preferences among companies.
    • You may politely decline to answer, citing OCS policies that discourage revealing preferences. Alternatively, if you believe sharing your preferences won’t negatively impact your chances, feel free to disclose them.
    • If you finish your interviews early, you have the flexibility to visit other companies that shortlisted you but weren’t in your initial four:
  • It’s okay to decline further interviews once you’re confident in your choices.
• Time-Wasting Tactics by Certain Companies:
  • Some companies, particularly non-quantitative ones or those with large shortlists, employ tactics to keep backup candidates waiting unnecessarily:
    • They may claim there’s one more round of interviews, wasting significant time.
    • This can lead to missed interviews with other higher-priority companies.
  • This is a very bad practice and has harmed a lot of students. Deal with it smartly.

Personal Experience

• I received offers from QuantBox and three other quant companies. I chose Quantbox Singapor Office in Quantitative Researcher Role.
• I was also shortlisted by Tower, Quantbox, Quadeye, Atlas, NK Securities, Trexquant, Graviton, Atlassian, Goldman Sachs, and Ebullient.

Conclusion

Internship season is undoubtedly a challenging yet transformative experience. It pushes you to your limits, helps you identify your strengths, and teaches you how to adapt under pressure. From preparation to tests and interviews, every stage is a learning journey.

The key takeaway is that preparation and persistence are your best allies. Use the right resources, stay consistent, and approach each step with confidence and determination. Even if things don’t go as planned initially, remember that perseverance often leads to success. Be open to feedback, learn from your mistakes, and never shy away from seeking help or advice from peers and seniors.

Most importantly, stay optimistic. The internship season is not just about securing a role; it’s also about growth and discovering what excites you professionally. Celebrate your efforts, regardless of the outcomes, because the experience itself is invaluable. Remember, opportunities will always come if you stay prepared and motivated.

Wishing you all the very best for your internship journey! You’ve got this! 🚀