Mobile Tech Lead

I'm a Mobile Tech Lead and former iOS architect. After studying AI, I now focus on Harness Engineering: shared harnesses that enable AI agents to build dual-native mobile apps (SwiftUI + Jetpack Compose) with alignment, quality, and speed.

What happens when we share the engineering system instead of the application code? Introducing HealthRoutine, an experiment in harness engineering for dual-native mobile development, with shared specifications, architectural constraints, agent permissions, and automated verification.

Naive RAG retrieval relies on vector similarity alone. This fails silently when documents are semantically similar but contextually irrelevant. Re-ranking adds a second filtering gate that evaluates actual relevance.

RAG isn't a single pattern, but a family of architectures. And choosing the wrong architecture isn't a performance problem, but a design flaw.

A practical approach to implementing RAG with LangChain4j, from a mobile engineering perspective. The gap between the architecture diagram and the working system.

Underneath all the layers of modern AI sits a single architectural breakthrough, the Transformer. Understanding it is not an academic curiosity; it's a requirement for designing reliable systems.

max_tokens is not a tuning parameter but a reliability control surface. This approach models LLM output as a probabilistic upper bound derived from schema, tokenizer, and model priors.

Pure vector search fails silently on exact identifiers. BM25 misses semantic paraphrases. In mobile AI, combining both signals isn't a performance trick — it's the minimum architecture for reliable retrieval.