logo_dotodo

Product: Semantic Search Engine - Web Platform

Client: Dotodo Inc, India

Introduction

Dotodo is a distributed, large-scale, low latency platform that seeks to connect people wanting to do similar things. Its tagline is ‘Lets do Things together’.

If I want to buy a camera, and if the platform can connect me with someone else also wanting to buy a camera at the same time in the same city, perhaps we can go together which will allow us to get a discount. An extremely powerful concept that required cutting-edge semantic search implementation.

Dotodo is owned by an IIM Ahmedabad graduate who is also a serial entrepreneur with many successful companies and successful exits.

Atlogys was hired as the CTO (Chief Technology Officer) to architect and develop and deploy DotoDo from scratch. The Atlogys CTO’s hypothesized and then engineered Dotodo to be a Rails based noSql system with Semantic Intelligence running on the Amazon Cloud.

Screenshots

  • dotodo - social networking platform
  • dotodo -web 3.0 business platform

Features

  • Users to login and post 1 liner write todos
  • Others can vote, comment, like, dislike a todo
  • Users are shown similar todos from other users
  • Creation of a social tag cloud on bases of user’s todo
  • A ‘me-too’ feature – I also want to do the Same thing !
  • fb/tw integration
  • News feeds
  • Buddies – concept of followers
  • Search todos
  • System of Notifications
  • Users can message each other to connect

Technical Challenges Overcome

Firstly we had to get a lot of data i.e todo’s content into the system. Then we had to use NLP modeling and Machine Learning tools to semantically categorize the content. This required tuning models and grouping the data into buckets.

Secondly we had to fine tune and customize Solr search engine to build our own customized search indices and algorithms.

A team of interns from MBA schools was also hired to manually classify the content initially which would then feed into the NLP models for learning and then progressive automation.

 

Impact

Launched in Delhi India. Had 1000+ users in the first 2 months. Many successful stories like

I was hungry and the platform connected me with someone wanting to goto Mc’donalds – so we went together 🙂