The Ultimate GGUF Guide: Why it became the de facto standard for local LLM inference
A practical deep dive into GGUF architecture, quantization patterns, and deployment tradeoffs.
GGUFOn-Device AIQuantization
Read article
Writing
Writing
Technical writing about on-device AI, architecture, App Review, and product engineering.
Running Gemma 2B on iOS: Reducing Metal shader startup from 10s to 1s
Runtime startup strategy and practical iOS inference optimization.
iOSOn-Device AIMetal
Read article
Why I moved SwiftSuDoKu from CoreData to SwiftData in production
Concurrency-focused migration notes and reliability outcomes.
iOSSwiftDataCoreData
Read article
On-device models vs cloud APIs: Cost and latency from a real iOS app
A practical look at retention, inference cost, and interaction speed.
On-Device AICloud APICost Analysis
Read article
AI wrapper architecture: Stop using monolith patterns for streaming products
A resilient stream architecture pattern for AI wrappers.
ArchitectureBFFAI Wrapper
Read article
Prompt engineering for 2B models: Think like a compiler
Constraint-first prompt design for small model stability.
Prompt EngineeringSmall ModelsGemma 2B
Read article