In our weekly column, we spoke to Amardeep Vishwakarma, CTO at Shine.com, India’s second largest job portal to understand the growth mindset he has applied to build a high performing team and the recent AI moves made to solidify its position in the market. Vishwakarma is a long-time InfoEdge (Naukri) veteran who has over a decade of experience in leading recruitment platform. At Naukri, Vishwakarma built the flagship recruitment software, a news source indicated.
In his present role, which he took over in July 2018, Vishwakarma’s focus area is to streamline search for job seekers and provide a seamless experience to recruiters. He is reimagining the recruitment software to shorten the recruitment cycle significantly and use AI/ML to provide improved user experience. “The entire recruitment process takes almost 70 days. It’s cycle of 70 days from starting to look for the candidate to bringing him on board. What we want to do is apply machine learning to shorten this cycle,” he said.
Since he took over, Shine.com has made a concerted effort to introduce AI capabilities to move up the ladder and be known as an AI-first company. Last year in December, Gurgaon headquartered jobs portal introduced face recognition and touch ID capabilities in its mobile application. The latest features rollout was aimed at enhancing user experience on the Shine.com mobile app by facilitating seamless login. This is also an indication of Shine’s focus on user-centricity and how it wishes to be a front-runner in the recruitment space by leveraging technology as a differentiator.
Vishwakarma, “The launch of this first-of-its-kind face and touch-based login features on our mobile app, well before any other player in the domain, further reinforces our commitment to simplify and adding value to the users’ interactions with our online platform.” He further added that Shine is also the only jobs app that is improving its hardware side as well to up the user experience.
Tech Stack @Shine
As a practice, Vishwakarma and his 40-member team leans towards open source technologies. “So, we basically are on Python. We use Python and we are hosted on Google Cloud Platform,” he shared. In terms of database technologies, the team uses MongoDB, MySQL. For caching and queueing, the team uses Aerospike or Redis. The tech team also uses RabbitMQ. Apart from this, we use Java for the backend, he shared.
Vishwakarma, an industry veteran, known for building teams from the ground up follows an innovative approach to scaling the team. Every year technology is changing, new databases come to the market, so it is very important for the team to stay updated. “We give assignments to the team with the new technology apart from the work and we just do a benchmarking or doing a PoC. We basically try the new technologies parallelly. We also have Shine learning where we basically ask our employees to take those courses and certifications to upgrade themselves,” he said.
One of the use cases discussed was applying machine learning to detect spam. “We are building models so that we can detect spam automatically as it is very difficult to check all the emails manually so this can be solved only through ML. We are also using the machine learning tools for recommendation,” he said.
Another use case is where the candidate is looking for jobs and the team gets to understand his preferences. So, the data is used to train a model to recommend jobs that match the needs of the user. “This is one of the core parts for us and we are building models around this so that we can better results to our job seekers,” he said.
What Developers Should Know Before Joining The Shine Data Science Team
Enjoyed this story? Join our Telegram group. And be part of an engaging community.
Provide your comments below
What's Your Reaction?
Richa Bhatia is a seasoned journalist with six-years experience in reportage and news coverage and has had stints at Times of India and The Indian Express. She is an avid reader, mum to a feisty two-year-old and loves writing about the next-gen technology that is shaping our world.