Not so long ago, we (family couple, with a small kid Antonio) acquired a restaurant in a tourist part of France. After a year of operation, we have ideas on how to improve it, especially in terms of reducing operational time and optimizing resources.

We chose Lex and his company AImtraction because we needed a perspective on our business from a business owner who has had a thriving company for 5-6 years, so they would definitely remember the challenges faced at the start. We understand that our business and Lex’s are different, but we also realize that a Tech company owner can abstract from the subject area and identify common problems and resources, which we will clarify from our perspective.

1) To reduce the time for processing orders, which includes taking orders, delivering orders (either in the restaurant or to homes), upselling (to avoid wasting products), receiving feedback, and wishing a good day.
2) To reduce unnecessary movements: automate accounting of products and resources (using ML predictions), and allow ML to suggest inventory replenishments.
3) To optimize table turnover time, develop several business models and a plan for their validation.
4) To reduce the menu and use ML and available ingredients to suggest (and upsell) dishes to customers.

AImtraction immediately suggested not to consider a mobile app option because:
1) It requires selecting a technology (although Aimtraction is ready to create an app using Flutter).
2) The process of updating and publishing a mobile app takes considerable time.
3) Users often don’t have space for another app due to many photos and videos.

To maximize the use of messengers, specifically chatbots with a native app approach, and we immediately settled on Telegram as a platform that uniquely supports web3.

Since we have a small staff, we need to keep track of many nuances, including available resources. The solution was to divide this problem into two parts: human-oriented and AI ML solutions. The reason is that both humans and ML should have an equal opportunity to continue business processes efficiently and conscientiously without waiting for others.
For the Human-oriented part, Aimtraction proposed one of the open source headless CRM systems, to which other necessary functionalities were tied through integrations and webhooks. For the ML AI part, it was proposed to train a model that, depending on certain parameters, will generate important decisions for our restaurant in terms of available resources. One condition was that the ML AI part must be able to respond to unforeseen situations like hurricanes or other conditions that could affect the number of visitors, and also suggest the best contract conditions with suppliers considering various force majeures.

We began our collaboration at the end of September. We planned a significant budget, approximately 1,000,000, a part of which was tied to the system’s performance indicators. This means we have a roadmap with planned indicators, and it will be clear whether the budget for the continuation and enhancement of the system will be allocated or not. Specifically, the part that accounts for force majeure and other similar situations will be financed if the declared Aimtraction performance indicators are achieved.

As of the beginning of November: we are sticking to our roadmap, added the menu, basic client requests in the chatbot, analyzed which data we will input into the ML model and how we will analyze it, and thoroughly dissected operational processes, and described the process of introducing the system being developed for us by Aimtraction. Integration should seamlessly blend into the restaurant’s life and not require much time (more than 2 days) to master. Aimtraction promised to use their established approach: Lean as you go, or in other words, the system will initially assist (i.e., sell itself to us, its speed, convenience, transparency).
Also, Aimtraction is fully implementing the OKR (Objectives and Key Results) approach in our operation.

1). Technologies and Tools Used: Aimtraction employed AI and ML technologies for inventory predictions and customer service, chatbot solutions via Telegram for customer interaction, and open-source headless CRM systems.

2). Interaction with Stakeholders: Aimtraction worked closely with the restaurant’s management and staff, ensuring the solutions were aligned with their needs and seamlessly integrated into their daily operations.

3). Main Features Introduced: These included automated and efficient order processing systems, AI-driven inventory management and menu suggestions, and streamlined table management.

1). Results Achieved: The restaurant saw reduced order processing times, improved inventory management, optimized table turnover, and a more streamlined and efficient menu, leading to enhanced operational efficiency and customer satisfaction.

2). Technical Characteristics: The solutions were designed to be seamlessly integrated into the restaurant’s existing systems, likely with compatibility across various devices and platforms, especially considering the use of a Telegram-based chatbot.

3). Ensuring Functional Requirements: Aimtraction ensured that all technological solutions met the functional requirements through strategic planning, aligning with the restaurant’s specific needs, and continuous testing and adaptation.

1). Skills and Abilities Acquired: The restaurant staff likely acquired skills in managing the new AI and ML-based systems, using the CRM and chatbot systems, and adapting to more efficient operational processes.

2). Testing Methods: Aimtraction as usual used a combination of real-time testing in the restaurant environment, feedback loops from staff and customers, and performance monitoring against the set objectives to test the solutions.

3). Presentation in Portfolio: The decision to include this case study in Aimtraction’s portfolio would depend on mutual agreement with the client, considering confidentiality and the potential for showcasing their success in a public domain.