Ayesha Andleeb@Pine_Calmuv6b

Using Vectara and Llamaindex we process tabular data using multiple models to get extremely accurate recommendations. Our solution can scrape product information from any e-commerce platform such as Amazon, Walmart, eBay, etc., and use RAG to incorporate customer-specific preferences to find out the best suitable products for price, features, ratings, etc. The solution facilitates the hybrid search mechanism i.e. keyword as well as semantic search capabilities. It also supports summarization for products e.g. concerning sales, trends, revenue, etc. The next steps are using a voice-enabled chatbot, and an alerting mechanism with product search across internet.

EZ-Read is a GenerativeAI comic book creation tool. It's primary use is for Creators who want to assemble a comic book using Generative AI for character, background images and book parsing with Generative AI summaries and metadata extractions, such as which characters are in which scenes. The tool was built for the hackathon in 3 apps (do to time zone differences) but will be a single platform ina future state. One thing we tried to forcus on was consistent character creation (in different poses). We got good results with Midjoureny, but there is no API. So we built our own using Stable Diffusion.

The CodeBlast Dream Catcher proposes an innovative approach to searching multidimensional space for knowledge based on the following eight principles: 1. An "all possible combinations space" exists in a multidimensional space where knowledge is discovered, not created. 2. This multidimensional space is best searched with LLMs using goals, as goals carry the recipes for accomplishing them. 3. There exists a multidimensional "all possible" Codestral goal space consisting of interconnected goals resembling a graph. 4. This multidimensional goals space can best be searched by remapping it to the 2D Infinite Canvas proposed in the LabLab.ai Build Your Business Startup Hackathon's "Navigating the Infinite Plane". 5. The infinite canvas can be created using a 50256 base number system derived from the GPT-2 tokenization labels. 6. To avoid the costly computational expense of base number conversion, hidden and unhidden states are created in the 2D infinite plane. 7. These hidden and unhidden states correspond to the conscious and unconscious mind, proposing that the human brain uses a similar mechanism to avoid the heavy cost of base number conversion. 8. Thus, searching for knowledge becomes a simple mapping problem in 2D and 1D space in both hidden and unhidden states. Business Value: The CodeBlast Dream Catcher approach offers significant business value through the following benefits: Efficient Knowledge Discovery Resource Optimization Enhanced Decision-Making Scalability Flexibility Strategic Advantage The CodeBlast Dream Catcher approach redefines knowledge discovery by leveraging LLMs and innovative mapping techniques to efficiently explore multidimensional spaces. By optimizing resources, enhancing decision-making, and offering scalability and flexibility, it provides a strategic advantage, making it a valuable tool for businesses aiming to lead in advanced AI knowledge discovery.