ignITe-programmet

Internship at AFRY I led a Proof of Concept for a Large Language Model (LLM) chat. The solution incorporated Retrieval-Augmented Generation, facilitating dynamic and interactive conversations with data. I did model experimentation, comparing Llama 2 and GPT-4, along with various embeddings models and vector storage solutions.Key Contributions:* Model Experimentation: compared Llama 2 and GPT-4, exploring multiple embeddings models.* Prompting Strategies: Implemented techniques to prompt the model for the retrieval of relevant results, optimizing user interactions.* Data Processing and Storage: Managed data processing and stored information efficiently in vector storage solutions.* Presentation: Communicated our findings and the proposed solution for a company.Working on a real customer project, focusing on visualizing real-time spatial data in a Flask web application.Project Highlights:* Back-End Development: Engaged in back-end development with SQL to manage and process real-time spatial data.* Front-End Development: Designed and implemented a user-friendly interface using HTML, CSS, and JavaScript for visualizing spatial data through a heatmap.* Validating spatial data* Optimization Strategies: Implemented sorting methods to optimize data presentation and enhance user experience.

Working as a preschool teacher. Responsible for Information and Communication Technology at the preschool. I have been involved in developing and presenting an action plan for digital tools in the preschool sector. This has given me the ability to immerse myself in new technology and analyze how it can support peoples learning progress. As the role as a Pedagogista I have been engaged in operational development. Planned the operational year, quality report, and was responsible for the content of educational evenings.