This Week’s BS In Data Science – #BSinDS

I saw an ad for a data science certification that said, “I became a machine learning expert in 10 months.” That’s BS.

Boot camps, certifications, and online programs are a great for a number of different uses. They can round out your data science training, kick off your journey, or help you transition into data science from another technical or research field.

The hyperbole and overselling are problematic. I get asked about these programs a lot. Aspiring data scientists look to these as a fast track into the field because that’s how they’re marketed. The truth is often far different. Anytime you’re thinking about starting a for pay program ask a few questions.

  • What percentage of graduates find a job within 2 months?
  • What is the distribution of job titles for those who find jobs?
  • What are the instructors’ backgrounds?
  • Who developed the curriculum?

Many programs don’t track student success. That’s a red flag. It shows they don’t care about student success. You want to see 60%+ of students get employment in 2 months. Next look at what jobs graduates are placed in. Are the majority settling for analyst or programmer roles? That’s another red flag. You want to see most placements in data science or machine learning roles otherwise the program’s not delivering.

Look into the instructors’ backgrounds. You’re looking for years of real world experience at top tier companies. Programs that source instructors from failed startups or straight out of school aren’t what you’re looking for. You’ll be taught partially with most of the exercises being insufficient to prepare you for a job.

I’ve been asked by colleges and programs to build a data science curriculum. I’ve turned them down because I’m not qualified. A curriculum is different from a topic outline or aggregation of syllabi. The process is complex because it involves a partnership between educational experts and subject matter experts.

You first create a curriculum that defines the overarching goals, high level course structure, and expected student outcomes at each milestone. Then experts in the field create courses to meet those criteria. Exercises and testing are validated against the curriculum to insure they will achieve the educational outcomes. Metrics are created to test courses and assess student learning goals.

That’s what you want to hear. Also having strong business partners involved in building the curriculum and courses is a positive sign. Business partners are looking to source candidates from the program so have a strong motivation to make the program successful.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *