Anirudh is an entrepreneur and engineer from Carnegie Mellon. He has built software and managed teams for 13 years, working extensively with machine learning and NLP. Apart from 3LOQ, he has also built other enterprise tech companies like KeyPoint Technologies and Cafyne.
While partnering with banks and telecom companies to improve their customer engagement rates through new technology, he realized that customers who cross a certain threshold of usage rates exhibited greater product engagement. This set him on the path to building the world’s first AI engine that builds habits, Habitual.AI. Habitual.AI is currently engaged with leading banks to enable habitual usage of their digital and mobile banking platforms.
Anirudh Shah: I got my first taste of data analytics as a career option when I joined the enterprise tech start-up KeyPoint Technologies as a software developer. I ended up transitioning to natural language processing and machine learning domains, which were new and esoteric tech applications back then in 2007. I loved the work and the rest, as they say, is history.
Anirudh Shah: The first large data set that I remember working on was when we had just launched 3LOQ. Our telecom client, a SE Asian operator, had shared call-data records for more than 2M post-paid subscribers covering a period of 6 months. This dataset was more than a 1000GB and was very rich as it had provided data points ranging from the phone used to location of the subscriber and the duration of the call.
Anirudh Shah: It was when we created a machine learning model that was able to predict the kind of establishment by observing the following data inputs (among other things):
Anirudh Shah: I follow Data Tau, Hacker News, KD Nuggets, Analytics India, Digital Vidya and Analytics Vidya.
Anirudh Shah: Tomas Mikolov, Abu Mustafa, Facebook AI Research (FAIR), OpenAI
Anirudh Shah: HPCC by LexisNexis, Spark, Hive, TensorFlow, Scipy/Pandas, H2O, Supersets
Anirudh Shah: We’re automating the process of building product habits with our patent-pending technology Habitual.AI.
Anirudh Shah: Performance is mainly based on:
Anirudh Shah: The open nature of Hive and Spark as well as the large investment by the giants (Google, Facebook etc) in these platforms is the real reason behind their swift adoption.
Anirudh Shah: Basic statistics and probability are must-have skills for data engineers.
Anirudh Shah: The BFSI, Telecom, e-Commerce and Content sectors have all the required data in place.
Anirudh Shah: For the BFSI Sector: Credit Risk, Marketing, Operations.
Anirudh Shah: Curiosity is important – understanding the ‘why’ behind decisions and solutions is key to professional and team growth.
Anirudh Shah: The real world is very messy. The sooner aspiring practitioners get their hands dirty, the better. Kaggle competitions are a great start.
Anirudh Shah: Make sure that you understand the basics of software development and approach data science by becoming a “full stack” engineer (full stack: exploratory data analysis, feature engineering, model development and optimization, production deployment, data integrity, ops, testing).
Are you inspired by the opportunity of Data Science? Start your journey by attending our upcoming orientation session on Data Science for Career & Business Growth. It’s online and Free :).