For someone aspiring to get started in data science, it might be overwhelming figuring out how to get started. You might be looking for that perfect formula that will transform you into a brilliant data scientist. However, there’s no such formula.
Everyone who works with data has a unique career trajectory. Some start as analysts and transition into data science, while others start as data scientists and then transition to data engineering. Even the way people get into data science—an online course, a master’s degree, opportunities within their careers—is unique.
To show you how everyone in the data universe has a unique story, we’re featuring interviews with successful data scientists. Last week, we spoke with Carla Gentry. This week, we’re featuring Ankit Rathi—Lead Architect (Data and AI) for SITA.
Meet Ankit Rathi—Lead Architect, SITA
We met Ankit Rathi while hosting a Kaggle Delhi meetup at our office. Ankit is Lead Architect at SITA, a leading specialist in air transport communications and information technology. Before joining SITA, Ankit worked in companies such as Genpact, Royal Bank of Scotland and Dell.
He regularly writes for publications such as Hackernoon, Towards Data Science, Analytics Vidya and more. He also shares his experience and learnings with young and aspiring data scientists through Medium and LinkedIn.
Read on to find out how Ankit got started with data science, how he wrote a book for data science aspirants, what are his go-to resources to stay updated and more.
How did you start your journey as a Data Scientist?
I started my career in 2005 with SQL programming. Since then, I have worked on various databases and tools related to data—from operational systems to analytical systems and dashboards. In late 2012, I read the now-famous article ‘Data Scientist: The Sexiest Job of the 21st Century’ and wondered what this new role was all about. I thought that I knew everything related to data by then.
While exploring the internet, I came to know about Coursera. It had launched its first MOOC on data science—Machine Learning by Andrew Ng. After finishing the course over weekends, I started looking for problems to solve to get some hands-on experience of working with data. This search led me to Kaggle, a platform for data science competitions. So that’s how I got started with the data science (DS) / machine learning (ML) / artificial intelligence (AI) field.
What are your go-to data science books or resources?
Blogs: Towards Data Science (on Medium), KDNuggets, Analytics Vidhya
Books: Machine Learning with R by Brett Lantz, Python Machine Learning by Sebastian Raschka, Introduction to Statistical Learning in R published by Springer New York, Deep Learning by MIT Press
Courses: Machine Learning & Deep Learning by Andrew Ng on Coursera, MS in Data Science by Harvard University (via GitHub)
Apart from these, I have learned mostly on the job via Quora and StackExchange.
What are the top 3 skills that matter the most to you as a human of data?
- Data-related skills (collecting, merging, understanding and analyzing)
- Software engineering skills (building pipelines and deploying models)
- Soft skills (coordinating with stakeholders to understand and maximize the business value of use cases)
Who are some of the ‘humans of data’ you look up to? What excites you the most about their work?
There are quite a few influencers who are doing great in their respective fields, mainly I follow:
Srivatsan Srinivasan: He knows the ins and outs of AI engineering. Since I also have a similar background, I find his articles and videos quite relevant.
Prudhvi Potuganti: He is an excellent educator in the field of AI/ML. Had I not been in the IT industry, I would also have been an educator. So I can relate to his work and I respect the work that he is doing.
Preksha Kaparwan: She is an excellent communicator. Explaining the complex stuff from the analytics field in plain English comes naturally to her.
I believe that I can also become a better AI engineer, educator and orator by following and learning from them, which will eventually help me as a professional.
Tell us about your work.
I worked on FlightPredictor, which is the flagship AI product of SITA. It starts predicting flight delays 24 hours before the scheduled arrival.
I can’t tell you much about the in-progress work, but what I can say is that we are focusing on infusing intelligence in our existing products. You can get more insights into the work we are doing here.
You have authored a book on DS and AI. Tell us about it. How did you get the idea, what is it about and who is it for?
Since the DS/AI field started picking up, every other day, I get at least 8–10 messages from aspirants and enthusiasts asking, “How can I get into DS/AI field?”. Over a period of time, I have improvised my response based on the follow-up questions they ask, such as:
- What is the difference between data science, machine learning, deep learning, artificial intelligence, data mining?
- What are the roles in DS/AI? Who does what?
- What concepts, processes and tools do they need to learn?
- Which books or courses they need to refer to?
At the beginning of 2019, I thought, “Why not build a learning framework to cover frequently asked questions and convert it into a book?” You can get more details about the contents of my book here.
What is a cool tool or library or a hack that you recently discovered?
Netflix open-sourced Polynote, the internal notebook they developed, to the public. Polynote could be another one of those great tools and the data science/machine learning industry need better tools to write code, experiment algorithms and visualize data.
I am still exploring it and I feel that Jupyter Notebook may now have a worthy competitor. Polynote is more like a simple version of IDE rather than a nicer version of a REPL. Let’s see how well the industry adopts Polynote. But definitely, it shows potential. You can read more about it here.
Describe yourself in three emojis and explain why.
😜 as I am quite playful and mischievous at times.
😂 as I love to have fun at work.
🤓 as I am a geek and a nerd.
Have more questions? You can connect with him on LinkedIn here.
Stay tuned for more such insightful interviews with the amazing Humans of Data. You can also subscribe to our no-BS fortnightly newsletter to get the latest interviews directly in your inbox. 📥