Software
- Successfully fine-tuned two LLM models: Gemma 2 1b and Llama 3.2 1b.
- Both models were fine-tuned using a Q&A dataset derived from Jules Verne's "From The Earth to The Moon."
- Started learning F' (F Prime) developed by NASA.
- Planning to use this framework as the foundation for the onboard software.
Hardware
No significant progress this week as I remain in the early exploration stage. Following recommendations from others, I plan to purchase an off-the-shelf CubeSat kit to begin testing and software development.