This project, initiated by a coworker and myself, involves building a system to simplify filing complaints for government services, such as reporting potholes. I focused on developing the frontend using SvelteKit and the backend with FastAPI. For the Large Language Model (LLM) component, we started with OpenAI's models and are now exploring the potential of running local models. A key feature of our project is the use of LangChain for the summarization of legal documents, such as a PDF book on laws, enhancing the system's capability to assist users effectively. The core idea is to create a user-friendly platform where people can easily interact with an advanced chatbot to address their civic issues.

At Velosimo, I developed UI components using React, honed team collaboration skills, and adapted to a production environment. I became proficient in Jira, Slack, and Figma for design mockups. My key responsibilities included building dynamic UIs driven by API data, managing state with Redux, transitioning the codebase to TypeScript, and implementing Jest for testing.