Note: We at Locale.ai are committed to making the geospatial industry more robust and mainstream. This is part 3 of the series “Ask an Expert” where we engage with geospatial experts 1:1 on interesting ideas, discuss geospatial techniques and tools and talk about the industry! You can read Part 2 here.
Pramod has been the Head of Data Science at Rapido since the end of 2019, a bike taxi service spread widely across all of India from Tier 1 to Tier 3 cities. Rapido is one of the fastest growing ride-sharing businesses in India. Pramod has had an extremely interesting career, from teaching at his alma mater- MSRIT, to working at ThoughtWorks to heading the Data Science Division at Rapido. He is widely acknowledged as a leading voice on the matters of Data Science, Micro-mobility and analytics.
We got together with Pramod to discuss his learnings about location analytics in India, the top uses of geo-spatial data in Rapido and his vision for the future. Let's jump right in!
(1) After graduating from college, you went back and taught Computer Science courses at the college. What sparked your journey from academia into the world of startups?
Pramod: I was always interested in academia from an early stage, around the end of 2010, along with a couple of friends and our HoD at MSRIT, we started solving problems associated with real life uses of Machine Learning in a lab. I had also been told that I was articulate and could explain concepts in a lucid manner. So, when my Professor told me that I should give something back to my alma mater, I took up teaching courses as a visiting lecturer. We had introduced a set of new age courses inline with what the industry needed such as ML, Data Science,Distributed System etc.
To be honest, there was no official transition period from academia to entrepreneurship. I have always been a vocal supporter of how organizations should think and act in a data drive manner, and I had also written at length about the same. While working at ThoughtWorks, I had the opportunity to work with multiple organizations to make their business more data driven, and while some organizations adopted newer practices and flourished, others did not seem it was worth it and I failed converting them to this direction.Co-incidentally, my good friend- Head of Engineering at Rapido, Srivatsa Katta gave me a call talking about the exciting journey which Rapido has embarked on.
At this point I had to take a call between convincing more organizations to follow a new way of life, and start to practice it on my own. I ended up choosing the latter, and I’ve never looked back since!
(2) What does a typical day in the life of the Head of Data Science and Engineering at Rapido look like?
Pramod: I have a very clear vision , philosophy and a goal as to how data will come to life at Rapido. To put it in a nutshell, the approach I follow, is to bring together a group of smart people to solve hard problems. The team should comprise smart people who have the correct attitude and should really want to solve the problems at hand by seeing everything through analytical lenses.My job is to paint the picture of what is possible, provide possible paths and provide enough clarity of direction so that we can all succeed in creating data driven outcomes. Seeing the bigger picture and knowing ‘why you’re doing what you’re doing’ gets you a long way in that journey.
(3) Why do you think geospatial tech is critical for mobility and ride sharing companies? How has Rapido integrated location analytics into decision making?
Pramod: When you look at the mobility industry- logistics, ride sharing, delivery- anything that moves- it is heavily reliant on geo-spatial things than any other industry. Humans cannot understand continuous entities and their behaviour, but are adept at perceiving discrete entities. For example, a sundial is continuous and hurts your head. On the other hand, digital watch is a discrete in terms of what is shown- this makes it much easier to read.
The mobility industry is unique in terms of how the unit of measurement is always hard to pin down due to it’s continuous nature because at every step of the way, there is some geo-temporal aspect at play. So the intent of analysing anything through a geospatial lens is to get closer to perceiving continuous things by making them discrete, and this is what businesses always need. It's all about making these continuous, sellable entities into discrete units that are measurable and understandable.
Geospatial data helps us understand what people want, when they want it, where they want it and how can we leverage this data into giving the best services and customer experience to them.
We’re also a platform for earning for our captains, so it is important to ensure that our Captains are happy with how we operate. That all-round satisfaction of all the stakeholders always leads to a successful business.
(4) India is very unique as a country when it comes to last mile analysis. What have been some of your learnings and how do you include those nuances in your decision making?
Pramod: This deserves a much longer answer, but I’ll try to provide a glimpse into it. Every country has her own nuances and trends. While designing a mobility service for London, you have to ensure that you feed the burgeoning underground network. For SFC, there is no extensive public transport system, so you automatically have to satisfy a large demand for transportation.
When it comes to India, the trend is very clear- India wants to move, and she will move, there’s no stopping her. Historically, we’ve seen this across many policy decisions as well. One of the pioneering efforts in this regard from the point of public policy is the Pradhan Mantri Gram Sarak Yojana- a multi pronged attempt with State and Central Governments pitching in to help connect India. This seed actually created an immense stimulus in the country under the belief that if people move, the economy moves.
In India, there is a massive desire to move- for work, entertainment or education-people want to move, and this has percolated beautifully into Tier 2 and 3 Cities.
Rapido currently functions in almost all Tier 1 cities except Mumbai and Pune to some extent, but we have a large presence in so many Tier 2 and 3 cities that we have seen this trend manifest wonderfully all across India. People want fast, smooth and cheap transportation- they don’t want to wait for a bus anymore. So, the greatest learning is that while India wants smooth and accessible transportation, unless you can address the affordability, you will not be able to sustain and grow.
(5) Geo-data is one of the most under-utilized data sets. What would you list as the top three use cases where Rapido has utilised geospatial data for business decisions?
Pramod: Recently, I attended a few workshops organized in Bangalore related to looking at mobility and transportation solutions from a geospatial lens. Some of us were gunning for a city where the visibility of the demand is much better orchestrated to ensure better serviceability. One of the questions asked at the forum was “Why do you want this greater visibility?”
To find an answer, let us look back at the conventional methods of transportation planning in a city. So there would be a survey floated around to gauge the top origin and destination pairs in a city, and a bus service would be started accordingly. But, what if this data set was available without a survey that is constrained in many ways? What if you can decide the best way to build these corridors just by looking at the demand data? That is what we do, at scale, at Rapido.
When it comes to our cities, we use geo-temporal data to understand what are the best routes, and how user behaviour varies along these routes. This ability to leverage location data gives us useful insights into what are the needs that we can meet, and this not only makes both customers and captains happy- but also gives us a competitive edge.
(6) Where do you see the geospatial industry going in the future? How do you think we can make it more “mainstream”?
Pramod: I think the most prominent and novel application of geo-temporal data in the future is going to be in the space of unmanned transport. Two years ago, the FAA (US equivalent of the DGCA) rolled out a concept paper and a draft proposal to discuss the management of drone traffic through data systems. This paper describes the future where unmanned micro vehicles will move through 3D space.
The important questions would relate to the creation of lanes, rules, and managing this traffic. I think the prominent role played by location data in this regard would be the creation of localised command centres that can govern and manage this traffic. The way I see it, location data will definitely be used to solve today’s problems of transporting people, and things through people- along with ensuring that this entire supply chain is smooth and efficient.
The most rudimentary application, of course, is to visualise demand and I think every business must pick up on it, if they are to survive. If not for anything else, do it for your customers and your partners.This is almost H1 for this space
In the future, the number of players in this field will be limited, but the scope of usage shall be vast. However, H3, the novel usage of geotemporal data in the future shall definitely be in the field of managing the traffic of unmanned vehicles.