Software Architecture Books, Papers & Projects

Everything here came from the same starting point: a problem that kept showing up in different companies, different stacks, and different teams, and a decision that the answer had to be architectural rather than incidental. The books, papers, and tools are all attempts to make that answer concrete and usable.

Books

Two books on software architecture and AI-native systems. The first, Universal Microservices Architecture, is published by Apress. It covers the full UMA model across 13 runnable chapters in Rust and WebAssembly, with 100% business logic coverage enforced in CI and a live reference application at universalmicroservices.com. It is the only book that treats runtime portability as an architectural primitive rather than a deployment convenience. The second, The Day After, is forthcoming from Apress. It builds the organizational model for restructuring software teams for the age of AI agents, role by role, using C-DAD as its structural foundation. Both books are grounded in the same conviction: that the problems most teams attribute to people or process are actually architectural.

White Papers

Four research papers published between 2023 and 2025, each addressing a specific structural problem that the industry was either ignoring or solving incompletely. CSMA (2023) applied service-oriented thinking to the browser and asked why frontend logic kept getting duplicated instead of packaged. UMA (2024) extended that question to every runtime in the stack and introduced the portable execution model. ECCA (2025) addressed event governance, closing the gap between schema registries and actual contract-level accountability for event-driven systems. C-DAD (2025) defined how AI agents can navigate software through machine-enforceable contracts rather than code. Each paper is freely available. Each one is also the foundation of shipped tooling or a published book.

Projects

Four open source projects, each built to prove that the frameworks work in running code. UMA-code-examples provides 13 chapters of runnable Rust and WASM implementations from the book, with every claim tested in CI. the-day-after-toolkit is a CLI that audits codebases for AI agent readiness and scaffolds the contracts needed to fix what it finds, configured out of the box for Claude, Cursor, and Codex. Traverse is a contract-driven Rust and WASM runtime for multi-agent orchestration scenarios, built spec-first before a line of implementation was written. youaskm3 is a personal knowledge layer that runs entirely on GitHub Pages with no server and no ongoing cost. All four are governed by OpenSpec contracts and CI-enforced quality gates.