Skip to content

I contributed fixed to gitcoin-web, giveth dapp, and undestrand the docker ecosystem very well.

I have been part of last EthIndia and improved on my skillset to create great HACKS!

Projects i am working on :

Map Usecases to a HTML control:
I am currently working a project that uses CNN neural net to map user stories from agile to generate a connection of HTML elemets representing a UI for the usecase, the connection generated runs on a framework that implements flow based paradigm.

We faced a mountain of roadblocks in the process, for instance there was a big problem of how to train the CNN and what kind of data should be used to train the neural network to solve this we had to create a intermidiate layer that generates html taking a graph of configs.

Another problem we faced is to bring the API and how to integrate to the framework, for that we used the swagger OpenAPI, we have written a parser that parses swagger yaml and generates API nodes that can be part of the graph generated by AI.

Cab Service:
I also have started a venture that targets vehicle owners and passengers, my attempt is to create a social network of driver and passengers which can eventually lead to a very convinient transportation system where rides with a driver he likes the most blocks a driver etc. This project is current is beta and running in my hometown with around 1k users and 50-100 drivers.

There are many problems we faced here,
We had to make it secure, profitable and affordable,
for security current we rely on regorus verification KYC for driver and users, In future we plan to create smart contract to maintain the ride history of each driver.

To make this app profitable for all drivers we had to fight with OLA and Ubers incentive system, we came up with 0 service charge idea. we won't be charging any cut ride wise from the driver they get what is the cost of the ride, where is the revenue comming ? its a secret ;)

Offline Synchronization:
I worked on a project that demanded to make a website run offline on all browsers when there is no internet connection and sync back any changes made when back online.

The project had a lot of challenge, one of the problem is to make it run on all browsers. service workers was out of the map with that. To solve this problem I had to use appcache for fallback on old browsers with a offline service that running on the system on a particular port. On detection of offline state the browser changes the base address from server to the offline service thus getting proper response in offline mode. Neat :) !

We also face ID problem, In offline mode all the Id of records create had to be changed on sync since it is out of order with server, that was a big issue we tried GUID but that made query and join table slow, To solve the problem, I used a base id for each offline service and all the records steps from there. Each service had to register a base address and a scope with the server before going offline and using ids from that scope making it unique and not used by any other service.