- Created a proof of concept desktop app that’s able to communicate with the keyboard using RAW HID reports. The desktop app is able to send a large block of text or data to simulate the context we are sending. The keyboard then types back a simulated summary, which is just the first and last words of the text block. This is to simulate that the LLM inference outputs in a stream, not act like an instant paste of the entire text block.
- The next step is to integrate this communication with Tatyana’s actual desktop GUI app
