Hustle Hub #27

New to Hustle Hub? Make sure to subscribe for more!

Hey friends,

This week I’ve been exploring OpenAI API (gpt-3.5-turbo) to build a secret project (more on that in the future 😜). If you haven’t tried out OpenAI API, no worries - I’ll talk more about it in today’s issue.

In today's issue, I’d like to share with you how to make your resume stand out using the STAR method, why you should be curious to try more stuff in college, the best ML book, and OpenAI API.

Let’s get to it! 🚀 

🛖 What's in the hub today?

• Tip: How to transform your resume with the proven STAR method

• Mistake: I only cared about grades in my college

• Learning: Be curious to try more things

• Book: The Hundred-Page Machine Learning Book

• Tool: OpenAI API

⭐️ 1 Tip

How To Transform Your Resume With The Proven STAR Method

Personally, I’ve been using the STAR method in my resume since the very beginning. It provides me a structured way to highlight my experience, and most importantly, to quantify my achievements to show the impact.

Most resumes I’ve seen tend to only list down what they did at their work but never show their impact. With the STAR method, you’ll be able to show your impact and capture recruiters’ attention to have one step closer to landing your dream job.

🧩 Situation

For example, here’s my work experience at Titansoft. I start by describing the situation or problem I faced at Titansoft. This sets the context for recruiters and helps them understand the challenge I was trying to overcome.

📝 Task

Next, I describe the specific task that I needed to achieve in the situation that I described. This helps recruiters understand what I was trying to accomplish and what was at stake.

🚀 Action

After listing down the tasks, I describe the specific actions I took to address the situation and achieve my tasks.

Be sure to use action-oriented verbs to describe your accomplishments, and focus on what you specifically did, rather than what the team or company did.

📊 Result

Finally, I describe the positive outcome or result of my action. Be sure to quantify your results using specific numbers, percentages, or other metrics (whenever possible). This demonstrates the impact you had on the organisation and highlights your achievements.

Because I didn’t have the number for the company’s profit margin back then, so I briefly mentioned the result without quantifying it (not ideal).

If you’re looking for jobs or planning to revise your resume, I’d strongly recommend using the STAR method to improve your resume.

This is important because employers are looking for candidates who can make a positive impact on their organisation, and the STAR method provides a structured way to showcase your ability to do just that. It can also make your resume more engaging and easier to read, which can help it stand out from other candidates' resumes.

In case you’re curious how the STAR method got me my first data scientist job, I made a video to walk you through how I wrote and structured my resume step-by-step. Just watch the video below👇🏻

⚠️ 1 Mistake

When I was a Physics student in college, all I cared about was grades.

So I studied day and night. My typical day was to go to lectures in the morning, and then tutorials in the afternoon. If there was a lab session, I might stay in school until the evening. Once I had my dinner and went back to my dormitory, I continued my studies to prepare for mid-term or final exams.

I didn’t join any activities or competitions that I was genuinely interested in, simply because I think these are not important for my studies.

Although I did well in my studies, I had always felt something was missing in my college life. I studied hard and got good grades, but I was still not happy.

Something was missing.

🧠 1 Learning

Looking back from today, I wish I could have tried more different things — like joining a startup hackathon (NTU Ideasinc) that I had always wanted to do, learning how to make YouTube videos, or simply building something to solve a problem that I cared about.

Instead of being a good Physics student, I could have learned more life and business skills that would benefit my future career.

🧠 Here’s what I’ve learned:

• Always be curious to try more things, especially during your college days.

• When you’re a student, you almost have zero risk to try things and build cool stuff that seems interesting and learn along the way.

• Once you graduate from college, chances are you’ll be less likely to try new things because you might have other financial commitments in life.

📚️ 1 Book

The Hundred-Page Machine Learning Book

This is one of my favourite ML books to learn the fundamentals of ML algorithms without being overwhelmed with technical details. It was written by Andriy Burkov — one of the best ML practitioners and thought leaders in AI in today’s industry.

📚️ Here are my takeaways from the book:

• Important concepts like linear and logistic regression

• Pros and cons of each ML model and their use cases

• Practical examples of how you can use each ML model

Have you read this book? What's your thought on it?

🧰 1 Tool

If you’ve been using ChatGPT recently, think of OpenAI API as ChatGPT on steroids. 🤯

That’s because you can build any Large Language Model (LLM) application by leveraging OpenAI API without building the AI model yourself - how cool is that!

That opens up a huge range of opportunities to build something useful for yourself or others, including:

• Personal AI assistant

• Build a landing page in seconds

• Work productivity tool

The best part? You can start using it for free as OpenAI gives you $18 free credits to get started.

For the beginning, I’d recommend starting with using the latest GPT-3.5 version (gpt-3.5-turbo) as it’s the most cost-effective one for their latest model. If you want to access GPT-4, you can apply and join their API waitlist HERE (I’m still queuing 😂).

🔥 Hustler Spotlight 🔥

👋🏻 How would you introduce yourself?

‍I’m Kevin Tran – a T-shaped individual who knows a little bit about many things and has 10+ years of experience in data science. I graduated with a Math and Accounting degree from UCLA in 2008 and M.S. in engineering management from USC in 2011. I have worked as an accountant, actuary, data analyst, statistician, programmer, data scientist, data engineer and recently technical lead in data engineering projects.

👀 What’s your day to day like in your past role as a Senior Data Scientist?

‍50% of the time communicating with stakeholders to learn and define what problems we are trying to solve, ask a lot of questions, and translate business requirements into a technical framework.

30% of the time get the right data, clean the data, explore the data, ask questions about the data, and iteratively work with the stakeholders very closely.

Once the problem is defined sufficiently enough (sometimes may require negotiation with the stakeholders due to lack of information), the data is relatively clean, and has enough information to proceed then the remaining 20% will be doing analysis.

Analysis here is pretty broad. It can be survey analysis, machine learning, statistical analysis, fancy visualization, Monte Carlo simulation, API development, invention of a specific algorithm to solve a niche problem in the industry, or deployment of a ML model to production. A typical project would typically involve multiple components of the list above.

⭐️ What has been the biggest highlight of your career so far?

Invented proprietary Monte Carlo simulation to predict the settlement value of multi-millions class action lawsuit - a first in the legal profession. The work led to a Data Driven Award in 2018 https://www.law360.com/articles/1105806/data-driven-lawyer-ogletree-s-evan-moses

Completed a competitive Data Science Bootcamp called The Data Incubator (4% admission rate)

As the only data scientist at LoanHero, I worked directly with a Chief Risk Officer on risk management and fraud detection accounting for over $100 million in total loan funding through the many projects. The startup was then acquired by LendingPoint in January 2018.

As a first data scientist on the Robo-Advisor team at Credit Sesame, I spearheaded multiple data science initiatives with the following highlight: Implemented and deployed a propensity model including A/B testing for validation. The model identified top 20% users with highest product interest scores who yield more than 50% of total revenue.

As the first and only data scientist in the Stanford IAIS team, I created an end to end technical framework (from survey design to survey analysis to Tableau visualization) to drive a data-driven approach. That approach has helped many other departments at Stanford improve their operational processes.

Recently switched over to data engineering and led multiple data engineering projects at both Facebook and Apple.

🚀 What's a data or AI trend you're watching this year?

‍Advancement in Natural Language Processing (NLP) such as transformers, ChatGPT etc

Data engineering toolings. There are so many toolings out there from so many vendors but most of them are doing many similar things.

💼 What advice would you give someone starting out in Data Science?

You have to continue to learn, and you have to learn how to learn. If you stop learning, you’ll become obsolete pretty soon, particularly in Data Science. These technologies are evolving every day. Syntax changes, model frameworks change, and you have to constantly keep yourself updated.

It doesn’t matter how smart you are, stay humble and respect everyone. Everyone can teach you something you don’t know. Treating people well, understanding their needs, and consciously working to see them as people instead of numbers or titles—this is how you succeed in the business.

To learn and grow, you must work with people, especially people with different skills and mindsets. Navigating your career is not all technical, even in the world of Data.

The thing that cannot be automated is having a heart.

Beyond this, you need a solid foundation. The one thing you can’t afford to do is take shortcuts. You have to learn the practicalities and how to apply them, but to be strong in theory as well.

Understanding what is happening underneath the code will keep you moving forward. If you take the calculator away, you still need to be able to do the work. You need the underlying skills too, so that when you’re in a situation without the calculator, you can still provide solutions.

🤯 What’s the most important career lesson you wish you’d learned earlier?

Thinking that after graduating from university, I know everything. Unfortunately no.

Thinking that after working in the industry for 10+ years, I know everything. Unfortunately, also no.

You have to stay humble and learn every day continuously. The more you learn, the more you will know that you don’t know a lot.

Learn fundamentals to build a solid foundation and don’t chase hype. Hype always comes and goes. For example, if you build yourself a solid programming foundation, it doesn’t matter if it is Python, Julia, C++, or whatever that language is, you can quickly learn new syntax and adapt to it. Remember, if else statement logic, iteration, variables, design pattern…are pretty much the same across many programming languages.

Stop wasting time on R vs. Python debate. Pick one and master one and preferably learn both because they are both good for specific tasks. If the project involves a lot of statistical analysis, I tend to use R. If it is data engineering, building pipeline, building ML model, I tend to use Python.

🧠 How would you learn Data Science if you had to start over?

Build a strong math and statistics foundation. Concretely, you would need linear algebra, some calculus, optimization, all the fundamentals of statistics. A house needs foundation to stay strong and it is exactly why you need a foundation to build a nice house–I mean a nice data science career. When the project gets complicated, internally you will start asking a lot of “why” questions. Your foundation will be your guide tackling those why. You will see this more as you gain more experience.

Build a strong programming foundation. Pick a language such as Python and master it. Look at the underlying source codes of many libraries such as Numpy, Pandas, and Sklearn etc. Be curious about how they implement functionalities that make your data science work easy. Start coding a lot and learn from your own mistakes by debugging.

🔥 Where can we find your amazing work?

Maybe follow me on LinkedIn https://www.linkedin.com/in/kevinidea/ 

I don’t have much time to dedicate to teaching others. However, I’m trying to do that. I have a personal project at www.kevinidea.com and this project may change any time.

Note that sometimes the server might be down but I occasionally check to make sure it is up and running.

📚 What's your favorite book?

• Second Edition: Machine Learning and Deep Learning with Python, scikit-learn, and TensorFlow: 9781787125933 

• Fluent Python: Clear, Concise, and Effective Programming: Ramalho, Luciano: 9781492056355: Amazon.com 

• Amazon.com: Machine Learning Design Patterns: Solutions to Common Challenges in Data Preparation, Model Building, and MLOps 

• Deep Learning for Coders with Fastai and PyTorch: AI Applications Without a PhD 

• Natural Language Processing with Transformers, Revised Edition 

• Data Engineering with AWS: Learn how to design and build cloud-based data transformation pipelines using AWS 

My own little math book: https://drive.google.com/file/d/0B-9Z0DcU7QACN2U0M2Q0MTYtYmEyZC00YzUxLTllZmUtNWJiNzJlNzQ1NGI2/view?resourcekey=0-_7OcRCZmck1kYNdX3KEYXw

🚀 Whenever you’re ready, there are 4 ways I can help you:

1. Book a coaching call with me if you need help in the following:

2. Promote your brand to 1,000+ subscribers in the data/tech space by sponsoring this newsletter.

3. Watch my YouTube videos where I talk about data science tips, programming, and my tech life (P.S. Don’t forget to like and subscribe 💜).

4. Follow me on LinkedIn and Twitter for more data science career insights, my mistakes and lessons learned from building a startup.

That's all for today

Thanks for reading. I hope you enjoyed today's issue. More than that, I hope it has helped you in some ways and brought you some peace of mind.

You can always write to me by simply replying to this newsletter and we can chat.

See you again next week.

- Admond