Conquering the 2024 Job Market: My Journey to Multiple DS/MLE Offers I — Job Search Summary and Strategy

17 min readMay 31, 2024

The last time I wrote a job search experience sharing post was on November 13, 2021. Time flies, and two and a half years have passed. I’m no longer the newbie in Zhongguancun, Beijing, but now an old newbie typing on a keyboard in a Yale University dormitory, preparing to graduate next month.

2021海內外數據科學、資料分析求職全攻略(壹) — 成績單、攻略、資源

How I passed 26 interviews & got 6 DA/DS offer from Disney+, Bytedance, SeaMoney, LINE TV, DiDi, Kuaishou — Part I

bertmclee.medium.com

At the end of 2021, I published a post sharing my job search experiences in Beijing, Singapore, and Taipei, experiencing a surge in traffic.

Introduction

In the past two and a half years, I have undergone a workplace baptism at Beijing’s Disney+ for one year and eight months, underwent master’s training in Statistics at Yale, and successfully landed a job again from the Asian market to the American market. I have decided to pick up the pen again to share my different experiences, insights, and growth in the DS job search. Hoping to help friends who still want to enter the data job market in 2024, specifically for:

Data work experience: NG (new graduate), career switchers, early career (1–3 years), mid-level (4–6 years)
Data job directions: DA (Data Analyst), DS (Data Scientist), MLE (Machine Learning Engineer)

Outline of the Job Search Series:

I. 2024 U.S. Job Search Experience Summary and Job Finding Strategies
* Introduction to the 2024 U.S. job search context
* 2024 U.S. Job Search Timeline & Results
* How to get interviews in the tough 2024 job market?
* Personal reflections and acknowledgments

II. How to Prepare for Product Case (& A/B Testing) Interviews (Meta)
* Meta DSA interview preparation experiences, frameworks, popular questions, and reference answers sharing

III. How to Prepare for ML Knowledge, Statistics, ML Design, ML Coding Interviews
* Compilation of past preparation and actual questions encountered in interviews (Expedia, AppLovin, Warner Bros, Walmart, etc.)
* Compilation of related preparation resources

IV. How to Prepare for SQL, Python Coding Interviews
* Compilation of past preparation and actual questions encountered in interviews (Meta, Warner Bros, Expedia, Shopify, CVS, etc.)
* Compilation of related preparation resources

Additionally, there might be writings about:
* Comparison of DS opportunities, job searching, and employment development in the U.S., China, Taiwan, Singapore, and the UK
* Comparison of DA, DS, MLE opportunities, job searching, and employment development

The past two and a half years have seen many changes worldwide, with significant shifts in global pandemics, economies, and technology. The employment market for data-related jobs has experienced considerable fluctuations worldwide — overall, there has been a significant increase in supply, while demand has fluctuated. Major tech firms like Meta, Amazon, Google, etc., have gone through crazy hiring sprees and severe layoffs post-pandemic. As of 2024, Tesla announced yesterday (4/15) layoffs exceeding 10%; Google in February laid off over a thousand people, including one of my strong friends; Amazon, Microsoft, eBay, PayPal, Snap, Expedia, and others also have had significant layoffs this year. Every day when I open forums and social media, it’s all about: “What should 2023, 24 new graduates do?”, “Sent a thousand resumes, received 3 online assessments, 1 interview, 0 offers”, “Job search anxiety, feeling a bit broken”…. This situation seems similar globally, as graduates in the U.S., Canada, UK, China are all facing considerable difficulties, a far cry from the times of massive hiring sprees with multiple big tech offers. When will it get better? I don’t know. For me, I dare not expect the future to improve; maybe now is the best time within the next decade. All I can do is what NCAA Women’s Basketball UConn star Paige said: “I did all I could so God can do all I can’t.” I do all I can, and leave most of it to God.

2024 U.S. Job Search Timeline & Results

At the end of June 2023, I resigned from my job at Disney+ and left Beijing. I arrived at Yale in mid-August. The Department of Statistics & Data Science at Yale offers two types of Master’s degrees:

Master of Science in Statistics & Data Science — This is a one and a half year program (August 2023 to December 2024), spanning three semesters.
Master of Arts in Statistics — This is a one-year program (August 2023 to May 2024), spanning two semesters.

Switching between these two programs is very easy, but it seems that all the students choose the 1.5-year M.S. program, which allows them to look for a 2024 Summer Internship rather than having to immediately find a full-time position, giving them more time.

Initially, I was also in the M.S. program. During the first semester, I was still contemplating life. Before starting the program, I had not prepared to look for a DS job, and I even considered switching to Sports Science, but I found the transition challenging and not very smooth. Later, as the coursework increased, I felt my math skills were too poor and struggled painfully, thinking I was not suited for DS/ML and considered becoming a PM, although I lacked the skills for that role, and even thought about becoming a doctor. Essentially, during the first semester, I randomly applied to perhaps dozens of DS Intern/Full Time positions, almost without any interviews, and just let it pass chaotically.

In January, inspired by who knows what, I decided to switch from the M.S. program to the M.A. and graduate early in May, suddenly reducing the time until graduation to just four months, and needed to fully focus on finding a Full Time position. At that time, looking at the bleak U.S. job market, I thought I might not find a job and was preparing to return to my country or move to another country.

Starting from January 9, I sent out applications and after a few days without any responses, I started binge-watching series (good ones and not so good ones), at one point feeling that the Data market had completely shut down, not even wanting Yale students. However, later on, I started receiving interviews one after the other. Below is the complete job search process flow chart and detailed timeline of job searching and interview schedule:

I applied to about 250–300 companies for 300–350 positions in total, received 18 HR calls, underwent technical interviews with 13 companies, reached the final/virtual onsite stage 6 times, and ultimately secured three offers, which are:

AppLovin — Machine Learning Engineer Intern (80–100% conversion opportunity after summer)
Expedia — Machine Learning Scientist II
Warner Bros. Discovery — Data Scientist II

Categorized by the content involved in the interviews, I experienced:

General HR Call: 18
SQL coding: 7 (Kafene, Expedia, Meta*2, CVS, Shopify, Warner Bros)
Python coding: 10 (①Leetcode: Expedia, AppLovin ②Pandas: Home Depot (Offline), CVS ③PyTorch Modeling: Walmart ④ML Implementation: Expedia ⑤OOP: Shopify ⑥General Coding: DataVisor*2, Warner Bros)
Product Case: 2 (Meta*2)
Statistics: 4 (Expedia, Meta, DataVisor, CVS)
ML Knowledge: 7 (Expedia*2, AppLovin*2, CVS, Warner Bros*2)
ML Design: 5 (Expedia*2, AppLovin*2, Warner Bros)
BQ (Behavioral Questions): 7 (Expedia, Meta, Home Depot, AppLovin, DataVisor, Shopify, Warner Bros)
Experience, Projects: Probably all, more or less.
Take-Home Project: 1 (DataVisor)

In this DS Job Search Series, I plan to share my preparation experiences, resources, and actual interview questions for the above interview items, one by one. Friends who are interested can continue to follow.

How to Get Interviews in the Tough 2024 Job Market?

Firstly, I must say that as an international student in the US, it’s really difficult to secure interviews right now. I don’t have the confidence to tell you that after reading this share you’ll have a more than 10% interview rate, or that you can get more than 10 interviews within a month. However, if anyone gets two or three more interviews because they followed my methods, then I think that’s enough. Perhaps those two or three interviews could change someone’s life trajectory, for better or worse lol.

First, a brief description of my background: I graduated from the College of Engineering at National Taiwan University, transitioning from engineering to data. After graduation, I worked as a DA for ten months at DBS Bank in Taipei and as a DS for one year and eight months at Disney+ in Beijing. Therefore, my identity in this job search is that of an early career Data Scientist with nearly three years of non-US work experience (actually 2.5 years), and because I went back to school, I can also be considered a NG (new graduate).

This time, the positions I primarily sought were DS and a few MLE/ML Scientist roles, ranging from NG positions requiring 0–2 years of experience (YoE) to early career, mid-level DS positions requiring 2–5 YoE, mainly focusing on the latter. NG positions seemed scarce, with each applicant easily numbering in the thousands, and I didn’t get a single interview — so my experience may not fully apply to NGs without work experience, but I hope it can still offer some insights.

Here’s what I wrote in 2021 about how to secure a large number of interviews:

Preparation phase: How to get a large number of interviews?
To summarize this topic simply:
1. A well-packaged resume.
2. Finding internal referrals.

How to write a good resume?
(---This section might not be very relevant to most readers, so it's omitted---)

How to find someone for an internal referral?
1. Maintain your LinkedIn, Medium profiles well: The opportunity with Twitter in Singapore came about because a senior from National Taiwan University noticed me on LinkedIn and offered to refer me proactively. Other opportunities, such as with Binance, Garena, McKinsey, and SeaMoney in Singapore, were introduced to me by HR or recruiters on LinkedIn.
2. Reach out to alumni, seniors, friends: I am fortunate to know people at Facebook, Amazon, Google, etc., in various countries. Some vacancies allow cross-border referrals, while others have restrictions. You can ask friends to check.
3. Follow communities on Facebook, Medium, PTT, LinkedIn: Many friends share job and interview experiences and help with referrals, like senior Junlin from TikTok Singapore, and friends from Shopee, SEA, and Agoda in Thailand who often share.
4. 一畝三分地(Forum): I found many referrals here, including for Disney+, Kuaishou, Ant Financial, Xiaohongshu, Pinduoduo, and others. These included opportunities in Asia-Pacific and the USA, though the latter required a visa.
5. LinkedIn Cold Messages: I got referrals through cold messaging on LinkedIn for companies like Ninja Van Singapore, GBD, and Line TV Taipei, but overall, the success rate isn't very high.
6. The blog '半路出家軟體工程師在矽谷': This blog is also a good reference.

In 2024, looking back at this share from the US, it is still quite relevant.

Before discussing how to secure interviews, I think it’s important to share a thought for consideration:

In today’s tough job market, it’s actually easier to compromise with oneself. It’s easy to think, “Well, I’ve already written my resume. I’ve sent out ten, twenty resumes today, so that’s it.” But in fact, most of the time, these efforts are ineffective. You really need to keep thinking: if there are tens of thousands of people competing with you every day, doing the same thing, what can you do a little more of, or do something different from others? How can you optimize your efforts? Otherwise, why should you win?

Now let’s talk about how to get interviews, still divided into two parts:

How to write a good resume?

This is a broad topic and hard to cover in just a few paragraphs. I’ll share some points, many of which are often discussed:

Resume Heatmap (from: https://mconsultingprep.com/management-consulting-resume)

Layout: Aim for perfection, make it comfortable and clear for the reader, focusing on what the recruiter will notice first, as recruiters may only spend a few seconds on your resume.
Grammar, Typo: Ensure sentences are smooth, without grammatical errors or typos. ChatGPT is your friend.
Content: What you did, how you did it, what the outcome was. A couple of things to keep in mind: 1. Highlight & Quantify your impact. 2. Think of your strengths that can help you stand out among thousands of resumes. What makes your resume different? Why should a recruiter choose your resume over others? You are not just writing a “good resume” that “clearly outlines what you have done”, but why your experience makes you an excellent candidate for the company.
Length: During my last job search, my English resume was 1.5 pages long, but after nearly two years of work experience, I condensed it down to 1 page — after several revisions, I found some things were repetitive, from a long time ago, or not so important and didn’t really enhance the resume.
ATS (Application Tracking System): Try to include keywords from the job posting (possibly some technical skills, models, business applications), as companies usually first filter with an ATS, and if it doesn’t detect the keywords, your application might be rejected right away. You might find an ATS Resume Checker online, like the one on the Yale Career Service website which can automatically score your resume.
Familiarize with the JD: This is similar to the fifth point above, the difference being that adding keywords here is to enhance the actual content of your resume. I suggest finding at least ten job descriptions (JDs) you are really interested in applying for, meticulously categorizing the positions (like Marketing DS, Product DS, Forecasting DS, Fraud Detection DS, Ads MLE, Business DS, BA, BIE, DA), then highlighting and finding keywords for each category, organizing high-frequency keywords, and matching your resume as closely as possible to those JDs. For example, as I applied more, I noticed for DS roles, some positions particularly want to see experience in Causal Inference, A/B Testing, User Segmentation, Personalization, Recommendation Systems, Marketing, Finance, Cross-functional Collaboration, Fraud Detection, Risk, Ads, Ranking, etc., and if you don’t have these, you might be at a significant disadvantage when applying for those roles; conversely, if you can package your experience around these keywords, you’ll have an advantage. Maybe check out some detailed JD analyses by the Little Red Book blogger [能量果汁工作室]: https://www.xiaohongshu.com/user/profile/62cee92d000000000200355a
Customize resume templates for specific functions: Since I initially applied for DA, BA, BIE, DE, DS, MLE roles, obviously each one is slightly different, it’s best to have a version of your resume for each function, and now that DS roles are very finely differentiated, maybe prepare different versions when you have time.
Customize for specific company positions: For opportunities I especially wanted to seize, I would base it on the function-specific resume template, and then let ChatGPT help me customize the Summary Section of the resume in LaTeX for that position, with the following prompt:

I found a job posting and need a standout summary in my resume, tailored to the specific job requirements. I'll provide 3 examples, the job description (JD) and my resume to ensure the message (in 3 lines, similar to the length of the examples) accurately reflects my real skills and experience, highlighting my 3 years in data science and relevant achievements. It's important the summary is attention-grabbing, considering the high volume of applications hiring managers receive, and showcases my suitability for the role without embellishing.

Examples:  \begin{rSection}{Summary}
\noindent Versatile Data Scientist with 3 YOE at Disney+, DBS and ongoing MSDS at Yale, excelling in Python, SQL, and BI tools. Proven track record in data-driven product development, analytics, and cross-functional collaboration. Ready to leverage advanced analytics and machine learning skills to innovate in Meta's dynamic product ecosystem.
\end{rSection}

\begin{rSection}{Summary}
\noindent Data Scientist with 3 years of experience, adept in developing scalable data pipelines and analytics solutions, transitioning to Data Engineering. Proficient in Python, SQL, and big data tools like AWS, PySpark. Proven capabilities in data processing and optimization, with significant achievements at Disney+. 
\end{rSection}

\begin{rSection}{Summary}
\noindent Yale MSDS candidate with 3 years at Disney+ and DBS Bank, specializing in ML/DL/NLP. Proficient in Python, SQL, and big data tools. My experience in user segmentation, coupled with skills in fraud detection and risk analysis, is well-suited for Stripe's data science roles. Eager to contribute my expertise and drive strategic insights.
\end{rSection}

[The JD here]

After writing a good resume, how to get interviews?

First, let me talk about how I encountered real people at the companies this time:

1. Kafene DE: Saw the Hiring Manager (HM) on Xiaohongshu posting about hiring for DE, so I contacted them.
2. Rize Education - Full stack DS: HR proactively reached out via email.
3. Expedia - ML Scientist III, Paid App: Applied through the official website + sent a LinkedIn cold message to the recruiter posting about the vacancy.
4. Stripe - DS: Saw the HM posting on LinkedIn and got referred by a friend.
5. Meta - DS, Product: Found an alumni on LinkedIn for a referral.
6. The Home Depot - DS, Marketing: Saw HM & Recruiter posting on LinkedIn, sent cold messages to HM & Recruiter, and got referred by a friend on LinkedIn.
7. Rippling - Sr. DS, Marketing: Applied through the official website.
8. AppLovin - MLE Summer Intern: Saw on Xiaohongshu that they were hiring, found a friend on LinkedIn for a referral.
9. CVS - Sr. DS, Patient Engagement: Found an alumni on LinkedIn for a referral.
10. CVS - Sr. DS, Product Platform: Found an alumni on LinkedIn for a referral.
11. CVS - Sr. DS, CMX: Found an alumni on LinkedIn for a referral.
12. 2K Sports: LinkedIn Easy Apply.
13. AvalonBay - DS: Saw on Xiaohongshu that they were hiring, applied through the official website.
14. Walmart - Sr. DS: Referred by a friend.
15. The Trade Desk - DS II, Forecasting: Saw Recruiter's post on LinkedIn, sent a cold message to the recruiter.
16. DataVisor - DS, Fraud Detection: LinkedIn Easy Apply.
17. Shopify - Sr. Product DS: Forgot where I saw they were hiring, couldn’t find someone for a referral, applied through the official website.
18. Warner Brothers Discovery - Sr. DS: HR reached out on LinkedIn.
19. Capital One - Principal DS, US Card: HR reached out on LinkedIn.

Summary of how interviews were obtained, categorized by method:

Referrals: 8 (Stripe, Meta, Home Depot, AppLovin, CVS*3, Walmart)/33 = Interview rate of 24.2%
Direct Applications: ①Website: 4 (Expedia, Rippling, AvalonBay, Shopify) ②LinkedIn Easy Apply: 2 (DataVisor, 2K Sports), 6/275 = Interview rate of 2.2%
HR Proactively Reaching Out: 3 (Rize Education, Warner Bros, Capital One)
Directly Contacting Recruiter or HM: 2 (Kafene, The Trade Desk)

In terms of how to apply for jobs, it can be divided into two parts:

① How to find effective positions:

First, regarding the choice of platforms, the more popular platforms for job searching in the US are LinkedIn, Indeed, Handshake, Glassdoor. It’s said that Indeed and Handshake have more positions at smaller companies because posting jobs on LinkedIn is apparently more expensive. However, I didn’t use Indeed because I find the interface ugly, and since I have been deeply engaged on LinkedIn for a long time, diligently managing my LinkedIn profile and having accumulated thousands of connections, I feel that LinkedIn offers me a greater advantage, hence I mainly use LinkedIn, but there are many nuances to using it. My personal priority for job application is:

Recruiters or HMs posting on LinkedIn Newsfeed that they are hiring; that’s a real active hiring sign. Some people compile posts they see on LinkedIn Newsfeed each week (not job board).
Direct hiring posts within One Acre Three Gourds community, actively providing referrals.
People on Xiaohongshu compiling daily new openings; apply to the freshest ones.
LinkedIn system-recommended positions, Xiaohongshu, One Acre Three Gourds; check the company’s official website for suitable positions to apply for.
LinkedIn Job Board, filter by Today/This Week, applicant under 100 is better; postings with 100+ applicants are usually not really hiring (green card purposes). However, the sorting on Job Boards is often poor, making it hard to discover suitable positions that get pushed back; better to search for companies you want to apply to and check their websites.
LinkedIn Easy Apply still has potential; I got two interviews through posts that were up for a while but had few applicants, probably because they were ranked low on job boards, unnoticed by most.

Regarding the issue of sponsorship, I eventually stopped worrying about it. When asked whether I need sponsorship when applying online, I would just say no and discuss it with the recruiter in person later. I met many recruiters, and almost none of them said they wouldn’t sponsor an H1B visa, probably because most of the companies I encountered are relatively large.

② How to effectively get your resume noticed by Recruiters/HMs:

Direct contact with Recruiter/HM. This mainly involves two methods: ① I spent a lot of time on LinkedIn sending cold messages to recruiters/HMs immediately after they posted job openings in their newsfeeds, along with sending connection invitations with notes. I also used some prompts to let ChatGPT help me customize invitation notes, but later I found this method to be less efficient. It seems this approach has been overused by others, and recruiters/HMs receive hundreds, if not thousands, of connection invitations as soon as they post, but it’s not completely useless, and it’s still worth a try when you have time. ② Another method is to find job search buddies in the same field to share contact information of recruiters/HMs, though I haven’t tried this method and don’t know how effective it is.
Referrals. I’ve already discussed how to find referrals.
Quick applications. Try to be among the first 100 applicants by using plugins like Simplify to speed up the process.

Personal reflections and acknowledgments

Since arriving in the United States last August, it’s been 245 days, or about eight months, which honestly feels quite short. I haven’t really had the chance to experience life as a student in the US, and my English is still broken and not confident. I still struggle to understand many menu items in restaurants, except for the English used in interviews which has become quite fluent, there is still much room for improvement in other areas. I’m still very unfamiliar with American culture and history, maybe only geography has improved a bit since I’ve actually visited some places. Fortunately, I’ve temporarily secured an opportunity to continue staying in the US to learn and grow, which I’m quite looking forward to, and I know I need to put in more effort to improve in various aspects.

During this job search, I experienced several minor breakdowns. Just talking about interview performance and outcomes, besides being disappointed about not getting an offer from Meta, being rejected by Stripe and The Trade Desk felt a bit inexplicable, and being rejected by one team at CVS made me question why (which led me to decline an interview with another CVS team as a form of revenge 😉, but on reflection, there are many possible reasons and deficiencies), and a few HR calls mentioned that my start date was too late or my years of experience were too short… still, I feel that everything went relatively smoothly or was acceptable. But after about three months of searching for a job full-time, with zero entertainment, minimal exercise & socializing, crazy resume submissions, LinkedIn networking, 18 HR calls, and 33 technical interviews, having over ten ongoing interviews to prepare for at times, I occasionally felt panic and exhaustion.

I am very grateful for the support and encouragement from everyone during this process, grateful for every prayer made for me, thankful to those friends who practiced mock interviews with me, shared interview experiences, and interview resources, and grateful to every friend willing to refer me. I am also very thankful to the many interviewers I encountered during the interview process; overall, most of the interview experiences were quite good and friendly, no matter what type of interviewer. Lastly, I thank God for His wonderful guidance, from National Taiwan University to Beijing, from Disney to Yale, and now from New Haven to the next destination. In every place, He has always treated me well, always looked after me, always given me far more than I could ask or imagine.

“But to him that is able to do far exceedingly above all which we ask or think, according to the power which works in us” — Ephesians 3:20