Inspiration

===================================================================

Let us think about pastime we are fond of. Travelling, fleeping news feed, watching movies, any more ideas? But what is not only pleasant but vital? The answer is food. We cook it ourselves, buy semi-finished products and visit restaurants. of course. Restaurants existed many many many years ago and are not going to dissapear. That's why our idea is connected with this service.

What it does

===================================================================

Imagine. You have just appeared at the restaurant door but the staff already knows, what kind of dish would please you most of all, while you have no idea what to order. What if you also get discount for your food preferences? We developed a system, combining features of: mobile apps for booking and searching restaurants; CRMs. DealWithMeal provides the further options: Automated process of a restaurant visiting (Searching by criteria(), booking a table) Intelligent recommendatory system, which helps to choose the most desired dishes. Discount system (for example, for favourite visitor's dishes)

How I built it

===================================================================

It was very complex process. We had a lot of work to do so it was very important to organize ourselves perfectly. We all know each other pretty well, so we had a good interaction and it helped us to get everything done in time. Finally, what was actually done? Firstly, we have spent a lot of time brainsorming and developing the main algorithm and architecture of our program. Secondly, we pointed the aims, that must be reached. It was enormous work to make the main architecture, think over questions about details in realization of some features. While we were implementing the project, we faced a lot of problems and bugs, some of which were so stupid, that we have laughed when they were fixed. Our system has the further main components: backend (Django Rest Framework), frontend (JQuery), iOS app, recommendatory algorithm (machine learning). Our algorithm is intended to advise dishes to client in 3 different situations:

  • It is client`s first visit to any restaurant (connected to our system);
  • Client has already ordered something in any restaurant (connected to our system);
  • Client has already ordered something in current order. There is no matter if client is ordering a table at website, by app or just at restaurant – we will advise him something he would be glad to eat! Our server is built using REST-architecture, so it has quite similar cases of interaction with app and frontend.

Challenges I ran into

===================================================================

The very first challenge was to generate idea. It is very difficult to come up with something new, but finally we've done it. And now we believe in DealWithMeal. The other challenge was to concentrate on work. Junction 2016 has manifold program: amazing events, interesting partners and guests, new people. But call of duty made us sit and code. We also managed to keep good communication channel between the members of our team. It is difficult to reach understanding, when everyone has a unique task.

Accomplishments that I'm proud of

===================================================================

Our team is proud of our project. It seemed impossible to perform such a result for 36 hours, but we dealed with it. We also dealed with developing entire, intelligent system, which can be really useful both for restaurants and visitors. We are also proud of each other. Every member of the team is vital, every member is inseparable part of Team "BORSCH".

What I learnt

===================================================================

First of all, we have learned how to deal with the problems if the time is short. Secondly, we have learned how to separate the duties effectively. We also have drilled the co-working skills and met many interesting people.

What's next for DealWithMeal

===================================================================

Our aim is to implement our project in real life. We are sure that it is very useful. We would like to extend our reccomendatory system and work not only with dishes, but with restraunts' logistics, purchasing; to improve the algorithm, to make it faster and more intelligent.

Built With

Share this project:
×

Updates