The following are the major inspiration for building OmniAssist app,

  1. Banking apps are dumb for tech-savvy & hardest for technically challenged. Banks has more than 60% of users from rural India who are not tech-savvy.
  2. Slower Digital Adaptation of Banking Services especially in developing countries.
  3. Technically challenged users are the primary target for social engineering frauds.

What it does

AI-Powered Mobile Assistant that can be invoked from any screen on user phone that has the ability to understand your financial context on the screen (like a financial conversation with a friend on Whatsapp etc...) and provide access to banking services in one click. While safeguarding the users from social engineering frauds.


  1. Easy Payment, Easy Referral, Easy Suggestions through context-aware payment assistant for banking services.
  2. Social Engineering Frauds Blocker through Social Engineering Fraud Firewall.
  3. Banking Queue Management + Banking Service Promoter through location-aware queue management.


We have designed this system to work offline and process in client-side, due to which our system is 100% scalable while not compromising on user's privacy.


If our system comes in production, we would be able to help more than 120M Indian mobile bankers to get adapted to online banking without being tech-savvy and help them from all type of cyber frauds and social engineering frauds, which causes a total loss of approx. $1B every year in India.

How I built it

  1. Frontend: Android
  2. Backend: Google ML, Firebase, Python, NLP
  3. Mock and test data

4. Finastra APIs USED:

a. Consumer Profile, Account Information(B2C), Payment Beneficiary, Person to Person Payment APIs used to retrieve user's bank account info and facilitate easy money transfer in one click by the Ghost Assistant.
b. Product Information, to retrieve the bank's product information to suggest to the users when the user enters the bank under the Bank Queue Management module.
c. Loans API used to recommend the user with loans availability whenever our context engine analysis user's need for loans.

Challenges I ran into

Training social engineering fraud module with the little bit of available data took much time than expected but was able to finish the POC in the given time.

Accomplishments that I'm proud of

I am proud that in a short time, with the help if Finastra APIs I was able to build an almost production-level and innovative product targeting the next billion.

What I learned

I never realise the ease of integrating financial services into our application and I always thought using financial APIs isn't for the common people/developer. Build as I started to explore Finastra API, I learnt how easy it to integrate the financial services.

What's next for Omniassist, Finance Assistant for the Next Billion

We want to add more assistive technology in finance and banking domain for our users through our platform, such as a. providing travel insurance automatically from their travel confirmation message. b. helping users to understand complex banking and finance jargon and many more.

Built With

Share this project: