Reflect is a physical AI assistant developed during an OpenAI hackathon. Designed as a compact, affordable hardware device, it aims to enhance daily life by offering intuitive, ambient interactions. Built with a focus on simplicity and practicality, Reflect is currently optimized for Espressif microcontroller devices, though the team hopes to expand support to additional platforms if the project gains traction. The device was inspired by several core principles: natural communication, seamless integration with everyday routines, and the ability to help users reflect on the past and prepare for the future. It’s designed to keep users in a state of flow by offering timely, context-aware assistance. Key features include location awareness, adaptability, and a low-cost design that makes it accessible to a wide audience. Reflect functions as a standalone smart assistant that operates independently of smartphones. It creates a local Wi-Fi access point named "reflect" when powered on. Users can connect to this network and access the device’s interface by navigating to http://192.168.4.1 in a web browser. Once connected, the interface displays two audio elements—one for real-time audio from the OpenAI API and another for audio output from the device itself—allowing for easy debugging and testing. The device is intended to serve as a quiet, always-on companion that communicates in a natural, human-like way. It can summarize the previous day’s events, offer gentle reminders for upcoming tasks, and adapt its behavior based on user habits and location. Its minimalist design emphasizes functionality and unobtrusiveness, making it a subtle yet powerful addition to any space. A video demonstration of the device is available, showcasing its setup process, interface, and real-time interactions. The project is open-source and welcomes contributions from the community. While currently focused on Espressif devices, the team encourages users to explore and extend its capabilities, with the long-term goal of supporting a broader range of hardware and use cases.