https://www.linkedin.com/feed/update/urn:li:activity:6371209812172201984
I began my data science career 15 years ago in 2003. First I learnt SAS. That lasted me till 2008. I then learnt R. That got me to 2015. Then I started using Python. In 2017 I started with PySpark. I love how data science means continuous learning , as learning is fun. We data scientists are also lucky to be in a high demand career.
- 609 Likes
- 23 Comments
- 84,446 Views
Expanded post by demand – https://lnkd.in/f7Ckk56
- 1,928 views
- likes 84
- Comment 2
- 3 shares
I began my data science career 15 years ago in 2004. First I learnt SAS. It was tough to learn coding as I had not done coding before in my MBA or my Mechanical Engineering. I remember coding from 830 am to Midnight one day. Most days I coded in SAS for twelve hours and ate lunch on desk. I used to remote submit code to Oracle datawarehouses and analyzed the then huge number of 150 gb. I did this both as an analyst in GE Consumer Finance Bangalore, and then as a manager in WNS Global Services. We were the first analytics team in WNS Knowledge Services and I transitioned projects for GMAC Insurance (arm of GM). I continued the journey with consumer finance data for Citi. The data pulled by SQL and analyzed with SAS.
That lasted me till 2008. I then learnt R. That got me to 2015. R was totally different than SAS due to object orientation. I shared my learnings in two books on R for Springer.
Then I started using Python. I didnot want my career to be dependent on one language only. Once again I wrote a book and did consulting assignments in both R and Python.
In 2017 I started with PySpark. It was fun but slightly different from Python. MLlib in PySpark was different from Sci-kit learn in Python for Machine Learning
I really really love how data science means continuous learning , as learning is fun. Yes I still enjoy coding.
We data scientists are also lucky to be in a high demand career. We get paid to have fun, to play Sherlock Holmes with data and get higher than average salaries.
Someone said Julia was promising. I checked it out a bit too. What is amazing is how SQL is still the most useful data science language. In 15 years SQL has endured.
2004- Present – SAS
2004-Present -SQL
2008-Present-R
2015-Present-PyData (Python Data Science)
2017-Present -PySpark (Python with Spark Big Data)
What will happen next 🙂
Author: Ajay Ohri
http://about.me/ajayohri View all posts by Ajay Ohri