Your DS Story S01E05: Aayush Rampal (Fractal Analytics)
‘Your DS Story’ is my attempt to bridge the gap between data science professionals & data science aspirants. Here new crop of data scientists will share their experiences, struggles, achievements & their advice so that data science aspirants/enthusiasts can learn and get inspired.
Aayush Rampal is an experienced Data Scientist with a demonstrated history of working in the management consulting industry. He is skilled in Apache Kafka, Big Data Analytics, Windows Server, and Apache Spark. He has been a strong engineering professional with a Master’s degree focused in Big Data Analytics from Shiv Nadar University.
1. Please tell us a bit about your background?
I completed my Engineering in 2012 from UPTU in Electronic & Communication. Though I always had interest in Computer Science and always loved coding since my school days, even then I wasn’t much clear as to what I wanted to take up as career. Luckily, I got to work in IT field as Analyst with HCL Technologies and I got to know a bit about Data Science while working on a project with Electrolux. My designation didn’t allow me to much that was in scope so I considered doing Master’s in Data Analytics.
Back in 2015, there weren’t much options and I was able to clear entrance for Shiv Nadar University’s Data Analytics program, one of a kind. It allowed to attend lectures for 1 year and 1 year industrial training. During my internship I worked at Tatras Data on recommender systems as well as Algoscale Technologies on multiple projects on latest technologies like Spark & Hadoop along with Machine Learning.
Currently I am following my dream to work @ Fractal Analytics on some really cool stuff like demand forecasting and what really excites my is how Artificial Intelligence can do everything what humans can do and so much more like driving cars and speaking Bots.
2. What projects you are working these days?
Currently I’m working on solving demand forecasting for CPG industry and its really important problem to solve as most of the companies have logistics and shipments in advance to demand they would face in future. So if we can crack how much demand for particular week/ month would be there for any vendor/store, we can help lower down storage and logistics cost.
Earlier I’ve worked mostly on unsupervised data and used NLP and Machine Learning hand in hand to solve problems like sentiment analysis, recommender systems and others.
I’ve used Object Detection in one of my projects solving how a pro-golfer plays and what angles can get the best shot to the target. In the same project we also classified type of golf club used by the players.
3. How your day to day job looks like?
Percentage wise I would say 10% is getting the data , 50% problem is solved if you are able to get right features and to get there you need to spend 20% time on data cleaning and visualization of features. Once you get right features, then its kind of cakewalk to do tuning as now we have very advance methods available for modeling and tuning hyper-parameters.
4. How you started with DS or transitioned into DS?
It was kind of hard to go back to academics once you’re already into industry, but education always pays and after searching for around two to three months, I was glad to get opportunity to work as intern . I always knew getting a job would be challenging so I took easier way out by going inside the firm as intern and then get a job once employer trusts you enough.
5. What advice would you like to give to DS starters or DS transitioners?
I would say its really awesome job and to get into it. Following are the courses which I think every Data Scientist aspirant should look forward to completing before starting for job hunt:
Stats courses on Khan Academy if they need to brush up.
Andrew Ng course on Machine Learning available on Coursera.
Machine Learning A-Z hands on Python & R course available on Udemy.
Above courses will surely help anyone understand basic ML skills, apart from this , one should never overlook coding and data structures part as that really helps while working on Feature Engineering/Data Pre-processing.
Start looking for internship to start with as only a few companies hire full time fresher data scientists and don’t really mind working even for free or lower wage if you’re getting some good projects to work on is one advice I have for all aspirants.
Thank you for reading my post. I regularly write about Data & Technology on LinkedIn & Medium. If you would like to read my future posts then simply ‘Connect’ or ‘Follow’. Also feel free to listen to me on SoundCloud.