About

The Project's Vision

The vision of this initiative is to help solve the problems discussed in the challenges with learning data science in 2022 section. This curriculum is a first step towards that goal.

Short Term

Broadly, the short-term mission is to democratize data science learning outside of the traditional education route by curating and organizing top resources from the internet. This is done in a few ways:

1. Macro level

   • The curriculum: manually curated list of the best possible resources from the web through rigorous content auditing for major data science topics. It also includes meta learning resources, and tips on how to get the most out of your learning experience.
   • Data Science learning path: an infographic showcasing what a possible learning path might look like.

2. Micro level

   • A resource hub: Sometimes there is incredible content aimed at explaining, visualizing and understanding specific concepts. Knowing they exist and referring to them in conjunction with a course you are already taking can really help you progress faster.

   Examples

   • This website that helps visualize probability basics
   • This small essay series by Andrew Ng on how to build a career in AI.
   • A community: TBD!

3. Open Source

   • Open Curriculai relies on contributions to help patch areas where content is lacking, and to stay updated with constant flow of new content being pushed on the internet.

All of the above are works in progress.

Long Term

• Ideally, every individual would be assessed in order to understand what they already know and what their objectives are. A personalized curriculum would then be specifically tailored for them which would evolve with ones preferences. Assessments could also be taken to track progress.

• People starting out in the field also need coaching, advice, guidance, support, and someone to help them view the bigger picture in order to reach their objectives.

• In light of this, tools can be created for smart, automatic content discovery, curriculum personalization, and cohort support and management.

If you're interested in helping out, have ideas about how to solve the issues, or just want to chat, feel free to reach out on LinkedIn.

FAQ

Curriculum Questions

Why choose a self-taught education?

• Take advantage of the abundance of high-quality resources available online. This curriculum includes many courses from top universities (MIT, Stanford, University of San Francisco), MOOCs and bootcamps with outstanding reviews (Deeplearning.ai, Fastai, Le Wagon), and content from world class creators in the form of blog posts, videos, and books.
• Focus on state-of-the-art techniques that you can apply in industry. There aren't many university courses or bootcamps that teach the latest techniques such as those found in Fastai or at Standford's CS224n Deep Learning for Natural Language Processing and then teach you how to deploy them to production.
• Have the flexibility to learn from anywhere around the world, and continue pursuing your studies part time if you find a job during the process.
• Become proficient in continuous learning. Following this path will teach you how to be autonomous in how you acquire new skills and rigorous in how you choose new learning material. These resources will accompany you at work and help you tremendously as you progress throughout your career.
• Put yourself out there and to showcase your work to others. Since you won’t rely on a college degree as a signal to get hired, you’ll be incentivised to create, share, and communicate your best work so that people will know what you are worth. These are essential skills when working at a company.

Does this curriculum offer a certification?

People looking for a traditional education and want a diploma should take caution. There are no certifications for completing this curriculum, just the individual ones you'd receive from the courses it lists.

What are the requirements for starting?

No programming experience is required and rough knowledge of high school math is expected. Also, for those who are committed to pursuing the curriculum end-to-end should already be extremely eager to learn data science, are self-driven and motivated. This requirement is crucial because a lot of the resources listed are self-paced. A high amount of discipline and will are therefore essential.

I am interested in a career in academia. Is this path for me?

The resources are chosen to prepare you to be up and running for an industry role. Learners interested in a career in academia and R&D should consider enrolling in a university instead.

How long will the entire curriculum take to complete?

This really depends on your existing background and if it is pursued full time or part time. Completing the curriculum end-to-end can easily take a year and a half full time if you have no prior background in data science. That said, the programme goes much deeper than a bootcamp and will give you more hands-on experience than most master’s degrees.

Won't I be disadvantaged during my job search compared to people graduating in data science with a formal degree?

While it is true that some companies will instantly disqualify your candidacy because you don't have a college degree, there are more and more people coming into data science with atypical backgrounds that have no formal education in data science. A strong portfolio and a resume showcasing all the requisite skills can go a very long way, especially as the demand for data scientists continues to outpace the rate at which universities can produce them.

But make no mistake, this path will require more effort, more dedication, and more self-discipline than if you were to enroll in a computer science or data science degree. While you can sign up for some courses & bootcamps during your journey to have some structure, most studying will be self-paced. You'll also need to spend the extra effort of thinking about and researching what to learn next (hopefully ODSC will help you on that front), and find ways to prove you have the knowledge required to be employed.

That said, taking this path is also, in our opinion, more enriching and fulfilling when you'll succeed. If done right, we believe you will be more qualified for an industry job than someone that is graduating from your average data science university program. Lots of employers will recognize the effort it took, and will give you a chance.

It is then up to you to demonstrate your skills and to convince your recruiters you have what it takes. Create a blog to write about what you learned and the projects you worked on, open source all the code you wrote on a platform like Github. Communicate about the grit it took to take a self-taught route and succeeding in it. When you feel confident in your abilities, start networking and reach out to your new connections about job opportunities.

Contributing

Can I contribute to the project and add a resource I find valuable?

Yes! Absolutely. This is one of the reasons why the curriclum is on GitHub.

There are two areas where you can make a resource request: 1. The curriculum, 2. The Resource Hub. Read about how you can contribute here.

That said, there is no guarantee that the content you submit will be approved. To keep a high level of quality of the project, and to avoid linking to too many resources, a detailed review of the content submitted will be done to either reject or accept it.

The Founder

Who wrote this why should you listen to him?

My name is Julien Beaulieu, and in early 2019, I decided to quit my job in digital marketing and engage full time in learning data science to transition careers.

All university programs I had looked at were first of all, not cheap, and second, due to start only 8 months after I was prepared to make the leap. I was not willing to wait that long to start learning so I enrolled in a few online courses. After several months, I realized the cheer amount of outstanding resources that were at my disposal and the potential there was to build myself a world-class education if I hand picked the best content. I just needed to arrange and organize everything in a way that made sense for my learning goals and interests.

Through the pains and lessons learned during my time studying, through auditing tons of courses, books, online content, by looking at other people's curated resource lists, and insipired by amazing open-source initiatives, I have developped this repo; something I would have loved to have when I first started.

Inspiration for the Project

How was this website built?

A very simple approach was taken to build this website: we used a theme called Material for Mkdocs, which is a simple static site generator initially geared towards building source code documentation. There are a lot of customization options available thanks to Markdown extentions.

The following website is completely open source and was an inspiration to some of the layout choices: Binbash Leverage.

All panda images you'll see on the site are generated by Open Ai. The website's logo was made using Midjourney.

Feel free to fork this project and view exactly how this website was made.

Credit where credit is due:

• The curriculum was inspired by OSSU's amazing self-taught, open source education in Computer Science.
• This website was inspired by Made with ML and BinBash who has open sourced its website's source code.