Four rounds of virtual interviews for Amazon SDE New Grad (VO)

Recently, I accompanied a friend to go through four rounds of virtual interviews (VO) for Amazon SDE New Grad, and the whole process felt like an upgraded version of the "intelligence + endurance" marathon. Amazon's virtual interviews are usually not conducted in one day, but are spread over two or three days, and this time they were spread over two days. The advantage of doing so is that there is a respite between each interview, but the disadvantage is that it requires two consecutive nights to maintain a high-intensity combat state, for friends in the country, this is definitely a double test of physical strength and willpower. Below I'll break down the interviews into rounds, and summarize the highlights for you.

Background of candidates

• Academic background: North American Master of Science in Computer Science
• internship: Worked as a software development intern at a well-known unicorn company, focusing on microservices architecture and API development.
• job search objectives: Seeking a SDE New Grad position with a large North American tech company, Amazon is an ideal fit.

This VO was scheduled during North American working hours and the four rounds of interviews were spread over two days. Although it was not a one-time competition, the continuous pressure in two days was quite high. Fortunately, my friend had made sufficient preparations before the interview, including several mock interviews, which made her play steadily on the spot and her rhythm was not disrupted.

Round 1: Coding + Resume Deep Dive + LP

The first round interviewer was a Chinese guy. The interview started with the usual self-introduction, followed by a detailed technical follow-up question about the internship project on his resume, involving design decisions and challenges encountered with microservices. He then asked two questions related to Amazon Leadership Principles (LP) Related issues:

• Ownership: Please describe a project for which you took the initiative and completed from start to finish.
• Insist on the Highest Standards: How do you ensure your code quality and project delivery standards?

After answering the behavioral questions, it was time for an exciting algorithmic session with two coding questions:

• Remove Duplicates from Sorted List II (Deleting Duplicate Elements in a Sorted Chain Table II): This question was handled relatively smoothly using the fast and slow pointer idea.
• Minimum Path Sum (Minimum Path Sum): This is a classic Dynamic Programming (DP) question, and my friend thought clearly and wrote the code quickly.

Overall, the first round was more of a warm-up, with a gentle pace from the interviewers, designed to look at basic competencies.

Round 2: Behavioral + OOD

The second round interviewer was a white American who communicated in English throughout. This round of interviews focused on behavioral issues and system design:

• Behavioral Question (BQ): The interviewer asked about the Bias for Action examples, such as "Have you ever made an important decision with incomplete information?" As well as examples of Deliver Results example, and follow up with the question, "In what situations do you have to compromise quality in order to deliver quickly?"
• Object-Oriented Design (OOD): The OOD questions for this round are Design a Vending Machine (Designing a Vending Machine). Focusing on an object-oriented perspective, my friend explains how to design various classes (e.g. Item, Coin, VendingMachine, etc.) and how to handle state transitions and exceptions.

This round of interview not only tests technical skills, but also communication skills. You need to express your design ideas clearly and logically, and be able to respond to the interviewer's follow-up questions.

Round 3: Bar Raiser (Pure BQ High Voltage)

The third round interviewer was an Indian interviewer, popularly known as the Bar RaiserThe interview was a great success. The entire round was focused on behavioral questions, and the interviewer had a very sharp style and liked to constantly interrupt and go after every detail in depth. It was very stressful:

• Tell me about a time you had a conflict with a teammate (Describe an experience where you had a conflict with a teammate): the follow up question is "How did you diffuse the conflict? What did you learn from it?"
• Describe a project that was a complete failure (Describe a project that was a complete failure): follow up with "How did you find out about the failure at the time? How will this experience affect your future work?"
• Dive Deep: Interviewers particularly emphasized Dive Deep This principle requires you to dig deeper into every technical decision and trade-off in your project.

This round is the most challenging of the entire interview. Answers must be drip-fed, and stories prepared in advance must be full enough to keep your composure under the barrage of questions from the interviewer.

Round 4: BQ + Coding

The final round interviewer was a friendly white woman. About 20 minutes before the interview started, both parties had a general discussion about BQ An exchange took place, which focused on the examination of LPs. She then gave a coding question:

• Reverse Words in a String III (Reversing Words in a String III) : This question asks to reverse each word in a sentence, but keep the order of the words the same. The question is clear and can be solved using either double pointers or splitting the string.

After these four rounds, my friend's energy was almost drained. But overall, he found the experience rewarding, as summarized below:

• code section: The difficulty of the topic is in the LeetCode moderate Around the same time, it mainly examines the impact on Double Pointers, Dynamic Programming, and Breadth-First Search (BFS) and other basic algorithms are mastered.
• OOD section: Often it will be classic design problems such as vending machines and parking systems, and the key is to make sense of object-oriented design ideas.
• Part BQ: This is one of the most important aspects of the Amazon interview. Especially in the Bar Raiser session, the kind of intense questioning that can only be dealt with if you're well-prepared in advance.

A few tips for those preparing for an Amazon interview:

• Emphasis on LP: Amazon interviews are heavily weighted toward LPs, and every principle needs to prepare at least one story that can stand up to scrutiny.
• Prepare multiple cases: Prepare 6-8 core cases covering a range of experiences, each of which should have the flexibility to be adapted to multiple LPs.
• Coding proficiency: When brushing up, you should not only be able to write the correct answer, but also be able to clearly explain your solution and code logic to the interviewer.
• Do more mock interviews: Practical exercises can help you familiarize yourself with the rhythm of the interview, adapt to the pressure in advance, and avoid malfunctioning on the spot.