Data Science is a very interesting field. On the one hand, the barriers to learning are low. You need only high-school math, and some python to get started.
According to the India Skills Report 2021 the highest skill gap during COVID-19 was in the areas of Data Science, AI, and Natural Language Processing (NLP). By some accounts there will be a shortage of 750,000 data scientists versus what the world requires by 2025. Traditional learning methodologies will not be able to scale in order to accommodate the rising number of trained resources. Part of the issue is the scarcity of teaching resources. On the other hand, the difficulty of becoming an effective data scientist – despite the myriad of courses claiming to turn aspirants into successful data scientists is contributing to instructor scarcity and the inability to churn out industry ready data science professionals. There’s a need to accelerate serious learning in data science and AI, making it available to a large number of aspiring learners. In a dispersed environment, there’s the added challenge of establishing talent and enabling employers and the talent to find one another. In an exclusive conversation with the Financial Express Online, Siddharth Das, Founder and CEO, Univ.AI talked about the situation of Data Science in India and scope of career attached with it. Excerpts:
Tell us a little about the talent gap that currently exists in the data science ecosystem?
Data Science is a very interesting field. On the one hand, the barriers to learning are low. You need only high-school math, and some python to get started. The field is also welcoming of diversity. To apply data science to solve problems in a variety of domains needs people who are skilled at both data science as well as their own domain. Contrary to popular belief, however, to become a good data scientist requires both ingenuity and experience. These are skills that take time to acquire. The need for both talent and experience makes data science a difficult expertise to cultivate. By implication, teachers themselves are scarce, and so are trained resources. I think the number of really high-quality data scientists in India is around hundred. This makes it very difficult for employers to even plan for hiring. Employers need to hire and then train.
- IBPS PO Recruitment 2021: Application process begins; check exam pattern, application fee, how to apply
- LIC Recruitment: Dates released for LIC AE, AAO Main exam 2021 ; check selection process, salary and more
- UPSC launches helpline for govt job aspirants from economically weaker section, backward classes
What is the roadmap for creating a successful community resource on data science and AI?
Our goal is to create a very high-quality community experience that lets you learn to measure yourself and also look for jobs and internships. We call it “Learn.Hack.Earn.”
1. Learn – We are releasing community versions of our live courses in a “cohort-based” flipped format – i.e. recorded lectures, and live labs and exercises. Starting early November 2021, community learners will be able to take a series of short, two-week courses and a series of one-day workshops that will take them from zero-to-proficient to the core elements of data science. For each two-week period, learners will be assigned into groups to learn together. Courses will be community-graded or auto-graded.
2. Hack – Hackathons every 6 weeks will let learners test their skills and publish their scores for consideration by potential employers. This is not meant to be a one-shot thing. We look at participation and progression. If you keep doing these hacks, then we look at a learner’s most recent performances. It also lets us assess progress over time. Hackathons carry cash awards for winners, the top ranker is named a Geoffrey Hinton Fellow (GHF). The entire initiative gets its name from this.
3. Earn – we are organizing leading employers to expose their data science programs to job seekers, to attend career fairs, and hire interns and full-time from this community. All of these programs will be continuous and ongoing, so that community participants can progress through the pipe at their own pace. That said, there is no parallel to this anywhere in the world, and we intend to make this a global recourse. We will initially make everything free of cost.
Highlight some emerging data science and AI trends and opportunities one can expect to see in the near future?
For one, no field is going to be spared. Every domain is slowly being upended. For example, AI can and will entirely replace control systems at process plants. We already know of the myriad applications of data science in medicine from diagnostics, to drug development. From shopping recommendations, and personalization to oil exploration and drone ops, we will see AI everywhere, somewhere enhancing human ability and in others, replacing it. Those trained in AI and data science will find increasing opportunities at the top of the food chain. On the other hand, certain white-collar jobs, such as entry-level programming might be at risk of being replaced by AI.
Please share some details of the upcoming launch of the Geoffrey Hinton Fellowship (GHF) and a participant’s journey to successful employment?
A launch is planned around October 20, 2021, right after the holiday season. We will release our first set of courses in early November, A participant’s journey through the system doesn’t have to be linear. Of course, if they are completely new to the field, they can begin with the first course. Otherwise, the course offerings will be completely modularized and they can take what they need. Over time we will add diagnostics to help folks decide what they need to learn. Hackathons are open and people can participate anytime and assess themselves, and apply for employment opportunities based on scores. If they do not qualify at the start, we expect that as they progress and improve their performance, they will eventually qualify for internship and employment opportunities.
Who is eligible for enrolling for GHF and what benefits can one expect?
Anyone can enroll. The major benefit is a self-paced opportunity to train and seek employment in a way that is completely transparent. GHF addresses 3 major questions among data science learners – where are the jobs?, how can I get them? and how do I prepare for them? The learning experiences carry over a lot of the flagship attributes. Live Q/A, labs and office hours every week – many of them with Harvard and UCLA faculty; cohort based review of video lectures, and coursework derived from Univ.AI’s flagship programs. In addition there will be onboarding workshops for beginners as well as advanced workshops for the experienced. Everyone at different stages of their data science and AI journey can expect something of value at GHF. Not the least of which is a self-selected collective, which, unlike other collectives, is highly engaged and in active pursuit of their professional goals.
You are simultaneously building a consulting arm at Univ.ai. What is the thought process behind that and which sectors do you envision to impact?
We have a resource base that’s richly skilled and high volume, we see that as a unique and unparalleled opportunity to participate in the AI revolution not just as educators but also as AI professionals. Consulting is an interesting format to do this because it allows a diverse set of resources to work on very diverse and interesting projects. As it turns out, the industry will appreciate this as well, as the top data science resources today are hired away by the tech giants. Our structure will make high-end resources available to everyone else.