xu-xiang

Integrates Apple’s on-device FoundationModels framework for secure, private, and offline LLM capabilities in iOS applications.

Generates high-quality, long-form content that mimics specific brand voices while eliminating generic AI writing patterns.

Generates human-like, brand-aligned long-form content including articles, newsletters, and technical guides while eliminating generic AI…

Provides specialized architectural patterns and best practices for building scalable, production-grade backend services and APIs.

Implements scalable server-side architectures, API designs, and database optimizations for Node.js and Next.js applications.

Enforces modern, safe, and idiomatic C++ practices based on the official C++ Core Guidelines for C++17/20/23.

Optimizes ClickHouse database schemas, query performance, and high-volume analytical workloads using proven data engineering patterns.

Transforms source assets into high-impact, platform-native content for X, LinkedIn, YouTube, and multi-channel campaigns.

Optimizes file processing performance by using SHA-256 content hashing for robust, path-independent caching.

Automates the capture and evolution of personal coding patterns into reusable AI skills through deterministic session observation.

Optimizes LLM API expenditures through intelligent model routing, budget tracking, and efficient caching strategies.

Implements safe, version-controlled database schema changes and zero-downtime deployment patterns across multiple ORMs and SQL dialects.