Big Idea

Automating Logistics

Machine learning start-up, Locus, is aiming for offline businesses after automating logistics for e-commerce players 

Photograph by RA Chandroo

"Any sufficiently advanced technology is indistinguishable from magic.” Nishith Rastogi, CEO, Locus, seems to have internalised Arthur C. Clarke’s insightful observation. His current goal is to eliminate all human decisions from the process of transporting a shipment from point A to point B. His long-term vision however, is much more audacious. He envisions a world where one day, humans themselves can be eradicated from logistics management, with the aid of machine learning algorithms, self-driving vehicles and mechanical arms. And if everything goes according to plan, Bengaluru-based Locus, a machine learning start-up that helps enterprises automate intra-city scheduled and on-demand delivery, will be a big part of that inevitable transformation. 

“Logistics today, is just data, data and data,” he says emphatically. “Every single rider of every single client of ours is a human sensor network. While we are talking right now, I’ve collected over a million data points,” he says. The amount of technology and automation required in logistics today is huge. “Now that we’re in, we have figured that this is a gold mine. Logistics is a problem suited for computers to solve. Nobody had solved it yet so people tried to simplify it using hub and spoke. But, it basically is high order equations that need to be optimised with hundred-plus variables. A human can’t do it, but a computer can.” To that end, Locus is an R&D driven, engineering-first start-up, with a small team of 25, 25% of whom are PhDs. “We will never be a sales-driven organisation, our valuation will depend more on our patents,” he says. They are currently in the process of filing three patents, a process which takes three to four years.

Early days
Their deep tech-focus is not surprising given that both Rastogi and his co-founder Geet Garg, both ‘hardcore tech guys’, met while working on machine learning at Amazon. However, the genesis of Locus lay in RideSafe, an app that Rastogi and Garg designed in the aftermath of the Uber rape incident in December 2014. RideSafe was a route deviation engine that sent out alarms if the driver was deviating from the set route. While it was an app aimed at securing women safety, the duo then di


You don’t want to be left behind. Do you?

Our work is exclusively for discerning readers. To read our edgy stories and access our archives, you’ve to subscribe