ShopperGPT is an innovative WhatsApp bot designed to be your personal digital shopper. Like a sophisticated chatbot, ShopperGPT is programmed to find the best product deals on WhatsApp and present them to you directly in your chat.
Technologies
Several modern tools were used to bring ShopperGPT to life. First, Langchain was used, which makes it easy to integrate language models with the application's backend in Python. This allowed for an effective and efficient implementation of the bot. Second, the OpenAI GPT-3 model was used, which is one of the most advanced language models available today. To add the ability to receive and process audio messages, Whisper was used. Finally, to get the best deals on the Internet, SerAPI was used.
Next steps: Set up an internal database management system with the aim of providing a more personalized and efficient user experience. Allow users to submit images of the product appearance they want and get satisfying search results.

StoryFaceAi is an online service for creating personalized photo gifts and decorations. It relies on advanced technology to design exclusive designs that meet the specific desires of each client. Users can choose from a wide variety of pre-made characters, specifying gender, ethnicity and age, to create a product that looks like them or matches the person they want to give the gift to.
StoryFaceAi's service is not limited to simple customization. It emphasizes ecology, using sustainable materials and adhering to environmentally friendly processes when creating its products. Whether it's a birthday gift, a holiday keepsake, or a decoration for your home, StoryFaceAi strives to create unique, high-quality designs while minimizing its environmental impact.

StoryFaceAi's proprietary technology and platform enables it to handle large numbers of orders every day, providing hyper-personalized service to thousands of customers. Thanks to this technology, each user can have a unique experience, choosing the design that best suits them. It is a scalable solution, able to adapt to an increase in the volume of requests without compromising the quality of service.

This project adds the most important functionality of translating messages between users. Also allowing you to configure different translation settings for each conversation thread.
Technologies
The OpenAI API is used for the translation service and the most important API to be able to capture and send text messages.

Next steps: The goal is to optimize the product so that it does not require configuration and is able to automatically detect when a translation is needed.

This project adds the most important functionality of being able to send audio messages as well as being able to send the transcript.
Technologies
OpenAI Whisper is used for voice recognition and transcription service and the most important API for capturing and sending messages.

Next steps: There are plans to optimize the product using OpenAI to optimize transcribed text by auto-completion of words and phrases. In this way, possible transcription failures are minimized to an almost non-existent point.

This case addresses the issue of the customer abandoning the relationship with the company. The goal is to predict and mitigate this phenomenon in a business using AI techniques.
Technologies
The machine learning is used to analyze large numbers of client data and determine patterns that can indicate that a client is at risk of abandoning a service. Other techniques of IA that can also be applied in this type of problem include the process of natural language (NLP), the mining of data and deep learning.

Next steps:
Deploy the model and perform other tests.

Customer segmentation dramatically increases business performance. Using AI, it is possible to group customers based on their buying habits, making it easier to personalize marketing messages.
Technologies
Customer segmentation can be performed using different technologies, such as cluster analysis that groups customers based on shared variables, business rules that establish segmentation criteria, machine learning models that identify complex patterns and interactions between variables, and visualization tools that identify behavior patterns and preferences. The choice of technology will depend on the context and the available data, so it is important to carry out a detailed analysis of the problem before selecting the correct option.

Next steps: Deploy the model and perform other tests.

The use of large language models (LLM) to generate coherent summaries and extract information is in high demand.
Technologies
Modern technologies such as OpenAI, Langchain, etc. are used.
Vector databases will also be used, these will suffice for the ability to store and retrieve vectors as high-dimensional points.

Next steps: Deploy the model and perform other tests.

Creating custom event flyers can increase recipients' engagement and interest in the event, offer useful and relevant information to them, and enhance the image of the brand or organization behind the event. Thanks to generative AI, it is possible to automatically create personalized flyers for events based on specifications provided by the client.
Technologies
Technologies like DALL-E and Stable Diffusion can be used to automatically generate creative content, such as images and custom text for the flyer.

Next steps: Deploy the model and perform other tests.

The Holy Bible is a multilingual website designed for people of all ages seeking a deeper understanding of sacred scriptures. This site offers illustrated Bible summaries in multiple sizes, allowing users to customize their reading experience based on how much time they want to spend. In addition, it has a unique feature: an interactive chat with an AI called GodGPT, which answers all questions related to the Christian faith. This site stands out for its customization for different age groups and the translation of all content into a wide range of languages.
Technologies
AI technologies used include Midjourney to generate images from text, OpenAI with Python to get summaries in different languages, and GodGPT, the chatbot, which uses a customization of ChatGPT with LangChain.

Currently, the project is under development. Explore the Bible in an accessible, personalized and multilingual way with "The Holy Bible"!

The Holy Bible Beta Test Subscription

GodGPT is a chatbot trained to convey the teachings of the Bible in a manner consistent with the teachings of Jesus Christ.
Technologies
ChatGPT, from OpenAI, is used as the base model, and for customization, LangChain, a framework designed to simplify the creation of applications using large language models. God_AI is also integrated with WhatsApp.

Next steps: GodGPT is in the development phase, it is currently a single-action agent, that is, for each question it gives an answer, but it cannot remember past information. A memory module is being implemented so that it can have a consistent conversation with the user.

Barbershop is an innovative project that combines artificial intelligence and 3D modeling to allow users to visualize different hairstyles before deciding on their cut. This revolutionary tool enhances the client experience and empowers stylists to communicate more effectively and deliver personalized service.
Technologies
The tools used for this purpose are Midjourney, a generative AI model for images, integrated with InsightFace, a 2D and 3D facial analysis project.

Next steps: We have a prototype of the service we plan to offer and we show some examples to illustrate the use case. Soon, we will add to the demonstration the possibility of using the service by sending a photo of your face (selfie) as well as an email where you will receive images showing your face with different hairstyles.
The Barbershop Beta Test

A text-to-image generation model takes text input and generates an output image based on the textual description. The model uses advanced deep learning algorithms to learn the relationships between the text and the corresponding images. These models are trained on large datasets of paired text and image samples to learn how to generate images that align with the given text description. The models employ a variety of techniques such as convolutional neural networks (CNNs) and generative adversarial networks (GANs) to generate high-quality images that accurately reflect the characteristics of the given textual input


Current Prompt:
Photo of a sunset behind the Eiffel Tower, shot 35 mm, realism, octane render, 8k, trending on artstation, 35 mm camera, unreal engine, hyper detailed, photo-realistic maximum detail, volumetric light, realistic matte painting, hyper photorealistic, trending on artstation, ultra- detailed, realistic In: ugly, low quality