Media & Marketing


Carlos Barge

It really depends on the company and industry. Whether it is data science or any other role, they are pretty well defined in a big or midsize company where an employee happens to do a very specific job for a long period of time. This happens because of the scale (and therefore hierarchy) of the work in a large company. It is not that you don’t get chance to develop your skills and climb the ladder.

In large companies:

It completely depends on the type of company.

a) In services and financial companies the data science work is mostly R&D type, you get lots of time to improve your machine learning competencies, work on data science projects of your choice, do experiments with different approaches, learn and keep up-to-date about latest in deep learning or some other cool algorithms. Most of the time the data you have access to is either very restrictive or less in size. You realize that you don’t need Hive or Spark just for 2–3 GB of data. At the end of the day, you ponder how are you impacting real customers.

b) In product companies, work is very fast paced (we call them “agile” and “sprints”). You hardly get time to do R&D stuff. More than your statistics knowledge, your programming skills will matter. Instead of learning newer algorithms you will learn about how to design scalable machine learning systems that can handle a million users daily. You get to work with different big data tools like Hadoop, Hive, Spark etc. Although at the end of the day you realize that your machine learning skill has not gone up by much but you are still content at heart that lives of your customers gets impacted by whatever you have developed.

In startups:

Startups are pre-dominantly product companies, so all things that you get to do in large product companies holds true in startups also, only that everything is more “agile”. In large companies, most of the time, you do not get the opportunity to build systems from scratch and you start to lose your cool looking at the legacy stuffs, whereas in startups, you often get opportunity to build ML systems from scratch. But building ML systems from scratch has its own set of issues:

a) When you are building systems from scratch, it implies that you don’t have enough data to work with at the beginning and that collection of data and how fast they arrive all depends on how good your system is, but most importantly, does your customers know about this startup and its product at all ?

b) Startups have very few employees and few resources at disposal. So it means that any labelled data you want has to be labelled by you or crowdsource the labelling task to other employees as well.

c) There might be very few backend and frontend developers and all of them would be busy with their own work. Which means most of the backend coding (API’s, Database, Caching etc.) as well as frontend stuffs (rendering images, CDN, AJAX queries etc.) has to be done by you.

Free Pre-Assessment Request

Do you want to know how your competitors are doing business?

Tell us a little about yourself below to gain data for free

Hi What’s your name?


Hi [First Name], what is your company’s name and website?



Is your company looking for any data on the following services:



Gotcha! Do you want to monitor any specific competitor or market?

List of Competitors

  • Add competitor…



Finally, what’s your email address and your phone number?



Your Data is on the Way!

Our data scientists team is working for you by collecting data and we’ll come back to you shortly with a pre-assessment and proposal.

WYgroup BI uses the information you provide to us to contact you about our relevant content, products, and services . You can unsubscribe from communications from HubSpot at any time. For more information, check out WYgroup’s Privacy Notice.

Leave a Comment:

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