Description
I'm looking for an experienced local-AI/LLM developer who can spend a few hours with me over 1 to 2 live screen share sessions teaching me how to run AI locally on my Windows laptop. I'm specifically interested in understanding the technology, not simply having someone install everything for me. I'd like you to explain what we're doing as we go and leave me with enough understanding that I can continue experimenting on my own afterward. My hardware includes an ASUS ROG Zephyrus with: - CPU: AMD Ryzen AI 9 HX 370 - RAM: 32 GB LPDDR5 - GPU: NVIDIA GeForce RTX 4070 Laptop GPU (8 GB VRAM) - Integrated GPU: AMD Radeon 890M - NPU: AMD Ryzen AI NPU - Storage: ~2 TB NVMe SSD - OS: Windows I'd like to learn how to make practical use of this hardware for local AI, particularly the NVIDIA GPU, while also understanding what the CPU, RAM, and NPU can and can't contribute. What I'd like to accomplish: - Evaluate my hardware and determine what size/type of models I can realistically run - Set up a local LLM environment (likely Ollama, but I'm open to your recommendation) - Teach me how local inference actually works at a high level - Show me how to download, manage, configure, and interact with different local models - Explain things like model size, quantization, context windows, VRAM/RAM requirements, GPU acceleration, etc. - Set up a user-friendly interface if appropriate (e.g. Open WebUI or something similar) - Explain how local models differ from cloud models - Ideally set up an autono