How this startup is changing regional news sharing in India with technology


Local is an Indian hyperlocal content app for non-English speakers in India. The application is available in more than 8 regional languages. In an exclusive interview with Analytics India Magazine, co-founder Vipul Chaudhary talked about Lokal’s services, technology stack and language corpus.

AIM: What is Lokal trying to solve?

Vipul Chaudhary: People residing in smaller cities have a different type of media consumption than ours. In small towns, people are more interested in what is happening around them because it affects their work and their livelihoods. But the small regional newspapers served them and did not do a great job. Newspapers ran mostly on government advertisements with no incentive to do a great job. So we had found a gap. Additionally, job portals and marriage platforms had virtually no matches relevant to the region and were forced to move to state capitals. So there was also this lack of information. We wanted to fill it and launched Lokal as an information marketplace for local updates, real estate listings, job searches, and more.

AIM: Tell us about Lokal’s offerings.

Vipul Chaudhary: Our core offering is relevant local updates. We also provide price information for vegetables, gold, silver, gasoline, diesel, etc. In addition, the platform also contains information about local health services and public services. In addition to this, we provide five classified services: Jobs, Matrimonial, Real Estate, “Buy and Sell” and Wishes. Wishes are unique to our platform, similar to the wishes we see in newspapers.

We offer our services in eight languages ​​including Punjabi, Telugu, Tamil, Kannada, Malayalam, Marathi, Gujarati and Bengali.

AIM: Which level corresponds to your main user base?

Vipul Chaudhary: Our main user base is customers in tier 2 or 3 cities. We also have users from Hyderabad, Bangalore and Mumbai. We have 20-22% penetration in one district, but it’s less in cities like Bangalore and Mumbai. Although we won’t focus on that, we believe that the higher level can also benefit from what Lokal offers in terms of local information.

AIM: Tell us about your linguistic corpus.

Vipul Chaudhary: Our regional linguistic corpus includes information that people have posted on our platform in their regional languages. We focus on how we use it. We try to take the information that comes to us on the platform and analyze and understand it through NLU. This is the input part. After understanding the language, we also need to understand the users. We mark their interests based on what they say. Now we have content understanding and user profiling; we take those two parts and then our job is to match them.

OBJECTIVE: Indic datasets are one of the main areas of interest in AI. How does Lokal collect regional data?

Vipul Chaudhary: Existing libraries and data are usually Wikipedia for NLP, or they have a body of existing data extracted from regional newspapers. Lokal generates 10 times more content than local newspapers. Our content talks about what is happening in the region, different topics and people. For local information we will ask where it happened, when it happened and what happened? So we have a corpus better labeled by default. If your input is better, your output will be better. We have a very clean, filtered, tagged input and a huge corpus to boot. So it helps us to build much more accurate models.

AIM: Tell us about Lokal’s tech stack

Vipul Chaudhary: Our technology stack has evolved rapidly. When I first built it, I used Python Django as a starting point because it could quickly get us up and running.

Even today, the core is Python Django. Our website is based on Node.js and React.js. Our Android app is based on native Android Kotlin. We have a parsing system we built in-house based on Red Panda and Apache with Kafka indices in the middle. We also have a lot of transcoding going on for videos right now. Our video transcode maker turns videos into streamable content because they are static videos. This service is also running on FFmpeg Codex, and for tags we use Python.

AIM: Given the magnitude of user data collected by Lokal, how do you ensure data security?

Vipul Chaudhary: We use the best standards to ensure data confidentiality. All of our servers are in India. We anonymize user data and analyze it in slices and slices. The machine would do a lot of analysis. We have completely isolated our production environment and our development environment on our cloud systems. No developer has direct access to our production environment.

All of our data is in a virtual private cloud to prevent outside hacking. Since the data itself is anonymized and isolated between environments, even if one can hack a system, they will not be able to extract personally identifiable information. In addition, the development of cloud security and data security is an active research area for us.


About Author

Comments are closed.