“It is very important to choose the right career path to excel.”
The world of the 21st century is well versed in all sorts of technologies. Such innovations have become a part of our life, via smartphones, televisions, etc.
The ease of operating such devices and the increased productivity can be accredited to the innovations in Artificial Intelligence and Machine Learning.
It is astonishing to see how incomplete it feels to live without the innovations of Artificial Intelligence and Machine Learning. Hence, we only expect more and more to cherish our luxuries.
Artificial Intelligence and Machine Learning are bound to prosper, especially when fed with properly equipped and skilled students.
Eduvoice Understanding The Needs And Bridging The Gap!!
Artificial Intelligence has grown exponentially in the past 2 decades and sees no signs of stopping. An important subject that has risen and is now a huge part of AI is Machine Learning.
Hence, we at Eduvoice, with the help of community power, wish to bridge the gap between the requirements of the corporate as well as research sectors and the college pass outs. The goal is to equip the students with the necessary skills to prosper in these sectors, via high-quality education.
Let’s Get To Know Mr. Chethan Danivas
We had the humbling opportunity to have a wonderful discussion with Mr. Chethan Danivas S. A, who is a Senior Professional of Artificial Intelligence and Machine Learning. In his current role, Mr. Danivas is working as a Senior Data Scientist at Avanseus. Let us dive further into his journey.
Mr. Danivas comes from a well educated and moderately traditional family. He feels that, by God’s grace, he was very fortunate to have an excellent environment conducive to good studies, expression of creativity, exploration, reflection, and overall development throughout – starting from his early childhood to primary, secondary, higher secondary education and beyond mainly due to his parents, teachers, and elders.
Mr. Danivas did his B.E in Mechanical engineering from Jayachamarajendra College Of Engineering. He was fond of mathematics and started developing an interest in computer science. During his final year B.E., he worked on a group project, under the guidance of B. Chennamalla Devaru, in the heat transfer field of mechanical engineering Although his interests were more inclined to computers, this gave him plenty of exposure to research methodologies.
Chethan later applied to the University of Texas at Arlington, Tx, USA for an MS in Mechanical. During his M.S., he also worked part-time as Research Assistant at Automation and Robotics Research Institute (ARRI), now called UTARI, under very eminent Professor Ramez Elmasri ( Author of “Foundations of Database Systems” book by Elmasri and Navathe)
Later, he made a switch to the Computer Science degree.
Mr. Danivas pointed out the difference in his experience of education in BE and MS. He told us, how in MS, they did not spend too many hours on theoretical education, and emphasis was laid on practical as well as assignments based knowledge.
The assignments were concept based and were given as individual work, and the projects would be assigned as either individual or group projects.
Chethan believes that students should be made to work by themselves instead of constantly telling them what to do.
He also informs us about the excellent lab equipment and faculty. In fact, the library and labs had 24 hours of access for the students. Mr. Danivas later came back to India and joined Seimens (SISL) Healthcare (HC) Group headed by Dr. Amitava Datta. Here, he worked as a senior software engineer and as a software architect.
Next, he worked for Infosys in the Finacle department (banking software development). All these roles gave him a lot of experience in programming, software engineering, software designing, and software architecture. Contrary to popular belief, he informs us how important programming is to get into AI and ML.
After working for about 10 years in the Software Design and Development area, he started his journey in Machine Learning in 2008. He was hence promoted to the research wing of Infosys, called the E and R wing. There, Chethan started working on image processing and machine learning to develop a facial recognition system.
He then started researching this field, and his mentor was Dr. Ravindra Babu, who helped him with his research. Mr. Danivas emphasizes the fact that mentors are very important for personal and professional development.
He developed various systems in facial recognition and he contributed to multiple US Patents, in image processing and pattern recognition, that the group was granted. Chethan also contributed to several publications.
Mr. Danivas then started taking courses from IIIT Bangalore. He has enrolled for a Ph.D. there but is currently on a break.
He later joined the Machine learning Centre of Excellence. (MLCOE) at Infosys. There, his role was to provide project consultation to internal teams of Infosys.
After his long and bountiful journey at Infosys, he started working with a start-up, namely Avanseus. Here, he is working on Natural language-based processing applications. They also work on equipment failure prediction.
A Word About Higher Education From The Industry
We spoke to Mr. Danivas about how the curriculum was outdated in our institutions. We also told him about the survey from Aspiring minds (2019), which stated that 80% of the technical field graduates are unemployable.
We explained how a lot of the subjects a student studies in college do not apply to our industry. Hence, we asked for his suggestion on how the corporate sector can help improve the curriculum.
Chethan suggested that colleges should look at industry trends and job trends. When the curriculum is being constructed, industrial experts should be called and their inputs should be taken.
He explains that if the curriculums are submitted to the Industry for review after being constructed, there is a chance of ego clashes taking place if corrections are suggested. Hence, continuous inputs from the Industry should be taken, so that the curriculum can be changed at the appropriate time, and not every year.
Mr. Danivas told us about how his mentor Mr. Laxmanan explained that there are two kinds of knowledge and skills. One type is the kind of knowledge and skills that are permanent and the other type are those that are temporary and changing.
Examples of such temporary skills are programming language syntax, specific API calling syntax of Libraries, etc. because these languages and libraries keep on changing.
Some subjects are permanent, like Machine Learning concepts, Machine Learning algorithms, Data Analysis techniques, etc, which are more concept-oriented.
Mr. Danivas explains to us that change in the concept-based knowledge takes place slowly because the research done in this sector is slower.
This is why the curriculum need not be changed frequently. Instead, the higher education sector should focus on building such foundational concepts.
The curriculum in Bachelors is not based on current researches, whereas in Masters, it is. Hence, the curriculum will not change as rapidly in the Bachelors’s education, as compared to the Masters, where it is highly research-based.
Developing and changing a curriculum should be well-coordinated with not only the industries but also the research field. The needs of industrial institutes like DRDO, known for Machine Learning, should also be considered while making such changes.
The research field will only progress if the students are employable for it, and for doing so; Mr. Danivas suggests the requirements of such institutes should also be considered while updating the curriculum.
He adds that this is one of the reasons for the lack of research in India because the research culture is not inculcated into education. He recommends that this culture should be inculcated and encouraged to do so from the bachelor’s degree itself.
Students pursuing masters should especially be encouraged to make publications and write books.
Chethan told us that we should pay a lot of attention to getting quality faculty for our higher education institutions. We need to have a mix of two kinds of faculty; one with academic background and knowledge alone, and another with additional Industry experience.
This is important because a faculty with academic background and knowledge alone would tend to develop the curriculum only from the academic point of view. Thus, a faculty with Industry Experience or research background is equally important.
He believes that if the person teaching does not have a practical orientation, then they will not be able to pass it onto the students.
Internships and projects need to be encouraged throughout bachelor programs because the students are made to work on projects only in the final year.
Mr. Danivas informed us that the students do have labs from semester 3 onwards, but they should be made to have larger-scale projects. The faculty should put in efforts to make the projects different for every batch.
For the labs, he advised that the Universities should have partnerships, with industries of the government or the private sector. Multiple partnerships with each of these sectors should be established, which has been done by IIIT Bangalore.
Even the faculty should be exposed to the industry or applied research. He suggested that the faculty should also be allowed to go back to the industry for 6 months or 1 year, and then come back and teach.
A notable point Chethan mentioned was that the institutes conducting the Master’s programs should take in students with practical experience. This in itself will significantly contribute to improving the quality of higher education. By doing this, the students will have a completely different approach to seeking knowledge, and they will know where this knowledge can be applied.
In some of the high ranked institutes, this criterion is already in practice. They require 5 years of experience before admitting students into management. Similarly, in technical institutes too, at least 2-3 years of industrial experience should be made necessary before the students are admitted to the master’s degree.
What Are Things The Artificial Intelligence and Machine Learning Fields Are Looking For In A College Pass Out?
Our next question to him was about the skillset a candidate should possess to function effectively and efficiently in the Artificial Intelligence sector.
Mr. Danivas told us that there were 2 aspects to this.
One of them is that the students should see this field as a career, not a job. This is a very important perspective to have because they must not simply aim for a high paying job. Such jobs may be high paying, but the learning opportunity is usually less at such places.
In a career, there are two dimensions namely breadth and depth. This is known as the ‘T’ model. Here, the students should focus on breadth-first (broad range of opportunities), and focus on depth in one or more areas.
This model is very much like the education system, wherein B.E. students cover a broad range of subjects and in M.S., students cover depth in subjects.
Mr. Danivas then informed us about the requirements of the private sector of AI but advised us that the requirements of the government and research-based institutes should not be ignored.
The organizations look for skills as well as knowledge.
In terms of knowledge, the students should know ML concepts. They should know the basics of machine learning, its theoretical aspects and limitations, and its applications.
Under the theoretical foundations, they are expected to know mathematics, probability, and statistics; basics of data analysis, linear algebra, mathematical optimization, information theory, and calculus.
On top of this, the students should know machine learning techniques, like supervised learning, under which classification and regression technique; unsupervised learning technique, like clustering and others like dimensionality reduction; re-enforcement learning, and neural network; which are the basis of machine learning.
The students should also have an overview of Artificial Intelligence. AI is a broader subject, and Machine Learning is simply a part of AI.
Mr. Danivas further explained that another important thing to learn is to handle and analyze different kinds of data, like structured data (like RDBMS table data), semi-structured data (like HTML) and unstructured data ( like free-flowing text, images, audio, and video ). ML processes and data visualization should also be learned.
Programming languages like R and python are essential for AI ML; and C++, as well as Java, will help them with building machine learning solutions.
He informed that other than this, the students should know common R and Python Libraries, other libraries such as MLlib and neural networks libraries such as Tensorflow and Keras.
Chethan tells us that students should try to implement machine learning algorithms from scratch on their own. This is important because if they only learn the tools and libraries, they won’t learn the nitty-gritty of these concepts.
They should be encouraged to write such programs. Thus, to be able to write a program from scratch is highly important for master’s degree-level learning.
Chethan also informed us that another must-know concept is of data engineering tools and libraries. For example, learning how to process files and data, how to prepare the data, is important for Machine Learning. These things come under data extraction or feature extraction. This is an important step in the Machine Learning Process.
He informs us that a fresher applying in AI ML will usually be asked to work in data engineering.
Apart from these, the students should be familiar with Big Data theories and cloud platforms, like Google cloud platform, etc. Knowledge of traditional and non-traditional databases is also crucial.
Next, we asked Mr. Danivas about the preference of the corporate when it comes to hiring students from tier 1, 2, and 3 colleges, and of varied IQs.
Chethan explained that the colleges may be able to provide education as per their rankings, but it is not necessary that Tier 1 Colleges will produce a more competent candidate, though the probability of finding more competent candidates in such institutes is higher.
Organizations look for candidates that possess the essential knowledge about Machine learning and the essential skills that are mentioned above.
He added that the second thing they look for is how well the students can apply the knowledge. It is not just a high IQ that is required, but also the willingness and ability to learn and adapt is what companies are looking for.
Other than technical skills, Mr. Danivas informed us that these companies also look for non-technical abilities like a positive attitude, how well the student can fit into the hiring team, group and the corporate, etc.
Another important skillset he informed us about was communication skills. As per him, these skills do not simply mean how well a person can speak or write, but also how well one can listen.
One should be able to put forth their point correctly and should be able to convince people. They should be able to analyze what the other person speaks before they respond.
Mr. Danivas told us that students should know when to use each machine learning technique and in what combinations. Being able to gauge the pros and cons of every technique as per the requirement is very important.
We then spoke to him about the hiring process in international companies like Microsoft and Tesla, where candidates are hired based not on the degree, but the skills they possess.
In India, a majority of companies hire based on the degree of the student. Therefore, we asked him if an organization in India will hire a candidate that does not have a degree but possesses the skills
To this, Chethan raised a very important point. He asked us how many people are skilled without a degree. Students may learn to program independently, but AI and ML, both are very demanding. The students need to have a degree and the right guidance to get into this field. This is more so important for freshers.
For people who are experienced, and already have a degree in any other field, the degree may not be as important, because the work experience counts. Here, the corporate might not have an issue. In fact, such an experienced person with a different degree will be preferred over a fresher having an AI ML degree.
Our next question to him was about a scenario where a student who pursued a non-technical degree, but has done projects in AI and ML and possesses equal knowledge equal to that of a technical background student. In this scenario, will the technical companies hire such a student?
Mr. Danivas told us that it is very difficult to replace formal degrees, and what is taught in the degree program, because it is very difficult to acquire knowledge at par with that, outside of the formal system. Also, providing career growth opportunities to such candidates, once hired, in the Corporate becomes difficult in the current corporate setting.
He added that companies are hiring students that have taken up B.Sc and BBA degrees and possess the relevant skills. He mentioned that students from mathematics background can also enter the ML field, they need not possess a BE degree.
Not only are these students eligible, but if they possess the required programming skills, they are also preferred for the role of a data scientist, especially in some foreign countries.
Mr. Danivas explained to us the importance of following one’s passion.
“It is very important to choose the right career path to excel.”
The student’s interests may change; hence he suggests that students should do a periodic assessment of their interests and try to realign/adjust their future career path, as necessary, with the changing interests.
He made a few more suggestions for the students. He said that they should attend the classes regularly and do their projects sincerely. They should learn more and more by making good use of the libraries and the internet.
The students should also work on professional networking while they are in the college itself. They should join academic and non-academic committees and student bodies, to work together and get their doubts clarified, etc.
He added that the students should take guidance from their seniors and other industry experts. They must be aware of the needs of the government and the private and research sectors.
A very crucial thing is that students should have good intentions and a desire to excel. Discipline, dedication, and sincerity are the torch bearers to success.
Mr. Danivas concluded the answer by talking about the vast career opportunities, in multiple sectors including Healthcare IT, Agriculture, Logistics, Supply Chain, Energy, Defence, Retail, Finance, and others, available in the field of AI and ML.
There are several industries offering jobs in the area of data mining, data analytics ( descriptive, diagnostic, predictive), Machine Learning/AI, and optimization based on these.
He wanted to address about analyzing the right data. If incorrect data is analyzed, the output will also be incorrect and useless.
Chethan brought to light several other opportunities, like in the finance department, where banks hire for the posts of cybersecurity and to prevent money laundering and data mining.
He added that even in the defense, there are many openings for the students, like manpower and air force prediction, logistics, injury protection, gauging the requirements of medical equipment and manpower, etc.
In the police line, crime prediction using data, face, fingerprint, and Iris recognition, etc. are some opportunities. Further, Mr. Danivas explained that airport security management requires video analytics.
There is a need for urban traffic management, traffic flow prediction, public transport need prediction, route optimization, even in air transport, these are required.
The institutes should give students local problems to solve projects. This will help the students learn and they will be encouraged to pursue a career in this field.
Most of the existing ML tools and libraries will help the students create a solution for these above-listed problems that arise as opportunities, but if these are not applicable, the students should be able to create and implement their own ML solutions from scratch by applying ML foundational and advanced concepts and suitable Algorithms.
“Our students should be trained very well, and they should work in Indian companies, and for the welfare of India.”
Mr. Danivas’s Say On Eduvoice And It’s Initiative.
Mr. Danivas felt that Eduvoice was a very good platform and we should continue forward on our journey. He advised us to consult the best and the most eligible people to get the best outcome out of the educational industry.
He added that Education should not be looked at as an industry, like other money-making Ventures. Education should be an Enabling System, performing an Enabling function in society and the Teachers / Gurus are the Enablers!
In the end, along with us, he wishes to bring about the welfare of India via education.
The broad objective of Education is
विद्यां ददाति विनयं,
विनयाद् याति पात्रताम्।
पात्रत्वात् धनमाप्नोति,
धनात् धर्मं ततः सुखम्॥
Credits:
Moderated By: Jayesh Pawar
Arranged By: Piyush Mohanty
For More Such Articles, News Update, Events, and Many More Click Here
