Moving Beyond Notion’s in Data Science/Machine Learning Domain

While I am about to complete my 7 years in domain of ML/DS/AI. It gives me pleasure to share my thought about variety of notion we have in this field. I feel it’s my humble responsibility to share my opinion and observations, as It might help aspirants as well experienced achieve bigger heights with less fuzziness in this area. Keeping my thoughts crisp and lucid, here are those:

1.??????No underestimation of simple tools: Learning a niche tool isn’t waste of time. We underestimate the power of tools like Excel. Humans, no doubt get fascinated toward new and jazzy products, and so we belonging to this fraternity get towards Python, R, PowerBI etc. But surprisingly, few things get way too easier, when we are well familiar with excel. Emphasizing again, let’s not underestimate power of simple tools.

2.??????Need of ML algorithm’s: Jumping directly to ML and Deep learning algo, without knowing the reason behind using it, is simply not appropriate. One must realize the need of using ML/DL and then use it. Knowing the fundamentals, the working principles of Algorithms, which kind of data for which specific algo used for, are things we should take care of. Just don’t rush for learning all algo’s in short span, go slow and but go deeper and stronger with basics.

3.??????Infer business sense: Everyone one thinks of ML and DS as area, where you apply certain techniques to get results, which can further be taken and presented to stakeholders. And this is where most of us fail to understand. DS/ML is not all about applying the techniques and getting the results. But it all worth when things actually make business sense.

One should always unhook himself/herself from data science and wear a hat of layman to see the real business sense coming out of analytics.

Data science can be taught, but business sense is somethings developed by yourself. The more we practice, the more we think, the more we understand the domain, all will add up to your business sense.

4.??????Automation mindset: Data science is complemented with automation. Always try to automate mundane and repetitive task. And while you in field of Data science, you will find ample of scenario around you, where automation will help you a lot.

5.??????Presence in digital world: Making your presence in digital media is key. Sitting in cocoon and doing modelling cannot help you in longer run. Alongwith, a good grip over technical stuff, your networking, attending conferences, writing into journals etc. can give your career new wings. It helps you grow holistically.

6.??????Get diversified exposure: Hopping organizations, a lot, may refrain you to get wider experience into data analytics project. A full fledge POC of Analytics project starts with lots of initial discussion with stakeholders, understanding data, fixing data quality, modelling it, making business sense out modelling results, presenting results to stakeholders. Each stage has its own skills requirement and things to learn. This all takes a good span of time.

?While I am not in strict favor of staying in just one org as well. Getting experience of different domain and organization’s environment can help you strengthen your technical and management skills. Dealing with variety of projects, peers, leaders bring flexibility and confidence in you.

7.??????Certifications: Certifications does gives sense of accomplishment. But if you don’t have practical skills, its next to useless. Emphasis should be on getting maximum project and henceforth practical exposure than certification.

8.??????Expecting smooth analytics journey: Keeping it straight, while there is no magic wand that can help you sail smoothly through analytics project. Analytics project journey is and will be bumpy. Getting to understand business, requirements, and data from stakeholders, experimenting with things makes journey frictional. But no doubt with time, everything falls into place. On one side, where Stakeholder understands the analytics and its inferences from various angles, other side analytics team also understands the business and its challenges. Finally leading to, coming up with optimized approach to solve business problems. Business and analytics go hand in hand.

9.??????Become avid reader and coder: Everyday reading acts as tonic to your brain. Start subscribing to some digital platforms or youtube channels where your mind channelizes the most. Read some few articles daily, it will help you flare in this domain faster. One may get innovative thoughts, idea and better way of solving business problems.

Few lines of coding daily, will keep your spirits up in this field. Coding is similar to math’s. The more you solve, the more dynamic and quicker you get in solving the problems. Coding does play a pivotal role in this domain.

10.??Drift from conventional learning methods: Rather than looking for best books(hardcopies), explore digital platform. Explore stackoverflow, towardsdatacience, Kaggle etc. These platform gives can you wide range of knowledge related to DS/ML. You may also look for various Youtube videos. One is expected to be updated with new happening in this domain always.

Disclaimer: Above mentioned views are purely from my experience, expected to vary from person to person.?

Happy Learning!