My book’s having more fun than I am. A book on a boat in Ibiza. It’s got a great view of the ocean on the East Coast, US.
It's making friends with Ginni Rometty in Singapore and hanging out with
@RavitJain
.
From Data to Profit has been more controversial than I expected and is
The Data Science learning path today is different than it was 3 years ago and looks nothing like it did 7 years ago. This thread has the main layers and example resources covering the basics, assuming you've got basic math covered.
1/18
#DataScience
#MachineLearning
If you're a Data Scientist who wants to be a better developer or builder, here's a thread on how to do it. There's so much bad advice out there, and I hope this helps clear things up.
1/8
#DataScience
#MachineLearning
#Programming
Dear data science job seekers,
The second you send a copy/paste DM or email asking for help finding a role, you’ve lost. Let me teach you a lower-effort framework that leads to more job interviews.
Python, SQL, Deep Learning, Docker, etc. doesn’t make a capable Data Scientist, but that's how companies write their job descriptions.
It’s like, “I want tomatoes, oregano, salt, pepper, basil, olive oil, and some penne” when you’re ordering lunch.
#DataScience
#MachineLearning
1. The data science tools, data infrastructure, and IT support they depend on aren’t there.
2. The budget and data team size are much smaller.
3. Specialized roles don't exist and data scientists are expected to play multiple roles.
It's telling that Apple and Google enforcing bans on hate speech is called a threat to free speech and the First Amendment. Free speech = accountability.
Creating a platform for free speech also requires safety. Removing hate isn't censorship. It says a lot if you think it is.
Continued with a lengthy video introducing data structures and algorithms in Python.
4. SQL. Must have because we still do so much in SQL, even with a top-tier Data Engineering team.
5/18
#DataScience
#MachineLearning
1. Research Methods. We do a lot of research and experimentation now. Data Scientists used to be model-centric but that's changed because our work must meet higher reliability requirements. I wrote an intro post:
2/18
#DataScience
#MachineLearning
4. The business isn’t data or AI-first and it has other priorities above innovation.
5. Internal users and customers aren’t data literate and technically inclined.
3. Software Engineering. We build for production now and our code must be as reliable as our models. It's no longer enough to learn how to write code. We engineer systems.
4/18
#DataScience
#MachineLearning
5. Data Visualization. I am pretty bad at this and it hurts to be bad at data viz. I'm constantly relying on help. Don't be me. This is a list of quality resources.
6/18
#DataScience
#MachineLearning
2. Causal Inference. Data Science has taken a hard turn towards causal inference, again to meet increasing model reliability requirements. An education on CI always starts with Pearl.
3/18
#DataScience
#MachineLearning
Change your approach to learning data science because the job you’re training for doesn’t exist. There’s no data scientist, just like there’s no software engineer. The term is a category. Target a phase of the data science lifecycle to build a realistic learning path. 1/5
If the job description asks for a minimum of 5 years of experience, it needs to include an explanation of why 4 years isn’t enough.
2/11
#DataScience
#MachineLearning
#Hiring
The hiring manager must spend as much time giving feedback to each rejected candidate as it takes for a candidate to apply for the job.
4/11
#DataScience
#MachineLearning
#Hiring
Take the skills list off your resume. Instead:
Tell hiring managers what you can do with Python.
Explain what you have built with TensorFlow.
Walk them through how you applied K-NN to build a solution to a problem.
#DataScience
#MachineLearning
#CareerAdvice
1. Spend a year coding as part of a team. Have people review your code and participate in code reviews. This will help you unlearn many bad habits. You'll also get exposure to different styles and best practices.
2/8
#DataScience
#MachineLearning
#Programming
6. Leadership isn't bought in and must be convinced before initiatives move forward.
7. Relationships and communications channels between engineering and the business haven't been fully built.
After 2 rounds of interviews, the company needs to explain what additional information they expect to get from this round and why they didn’t get it during the last round.
3/11
#DataScience
#MachineLearning
#Hiring
@sehurlburt
Hiring managers take note, job hopping can result from this cycle. Job hopping for women and minorities is often less a sign of reliability issues and more a search for inclusive, non-toxic environments.
8. Data Science Basics. I built this video with some of my favorite free resources for learning the basics. These are off the beaten path resources that give you more than generic content.
12/18
#DataScience
#MachineLearning
This job description framework gets qualified Data Scientists:
1. Inputs
2. Outputs
3. Outcomes
4. Supporting Teams
5. Customers and Stakeholders
6. Tech Stack
A skills list gets you 200 questionable applicants.
#DataScience
#MachineLearning
2. Build traditional software engineering type projects. Services and Web Apps are great because you'll learn fundamental coding skills.
You'll have to Google a lot which is a software engineering superpower.
3/8
#DataScience
#MachineLearning
#Programming
I was an aspiring Data Scientist 10 years ago. I had pitched the role and gotten several “No’s.” I started to doubt the role and the need, so I applied for more traditional jobs. My story can help you succeed now.
1/13
#DataScience
#MachineLearning
#CareerAdvice
Getting into Data Science for the money? Focus on strategy and building communications skills as well as the technology.
Why? My hourly rate for strategy consulting is 2X my hourly for ML Research and Model Dev.
#DataScience
#MachineLearning
#Strategy
#CareerAdvice
6. Cloud. I'm proficient with AWS and that's done well for me.
Azure is gaining in popularity but its search filters aren't as cool as AWS.
#DataScience
#MachineLearning
8/18
If you’re a senior leader reading this and thinking, ‘That’s absurd!’ Now you know how candidates feel looking at the hiring process.
6/11
#DataScience
#MachineLearning
#Hiring
Quick note. We're all lacking somewhere because the job is massive. I'm weak in data visualization but it hasn't kept me from being a Data Scientist. Don't feel like we're all masters of the full list because we are not.
7/18
#DataScience
#MachineLearning
None of these are realistic, but neither is the hiring process. The only way for me to make the strangeness of the whole thing obvious is through hyperbole.
5/11
#DataScience
#MachineLearning
#Hiring
Ask any business that’s successfully deployed
#MachineLearning
& they’ll tell you the hardest part is the last mile; from prototype to production. Path to production is a critical success factor for every ML project.
By the time you get here, you should have a good-sized portfolio of projects. A good portfolio has:
Analytics project
Research project
Data Visualization project
Implemented, running ML project
Write up each one with a blog post.
18/18
#DataScience
#MachineLearning
Data scientists must support the core business, deliver bottom-line impacts, or prepare for layoffs. The hype cycle is over, and results are all that matter.
An app that uses ChatGPT and DALLE 2 to generate slides from presentation outlines and Jupyter Notebook visualizations is the next $1B overnight sensation.