Writing

Blog

Technical notes on on-device machine learning, Apple's MLX framework, edge AI, Core ML and iOS engineering.

June 2, 2026·11 min
Running LLMs on iPhone with Apple's MLX Framework
Quantization, KV-cache and the memory math that makes on-device LLM inference work on Apple silicon.
MLXEdge AIiOS
May 14, 2026·9 min
Edge AI on iOS: Why On-Device Beats the Cloud
Privacy, latency, cost and offline reliability — the case for running ML where the data lives.
Edge AIOn-Device MLiOS
April 8, 2026·10 min
MLX vs Core ML: Choosing an On-Device Inference Stack
When to reach for each of Apple's two inference stacks, with a decision checklist.
MLXCore MLiOS