Skip to main content
James Hu
HomeBlogPortfolioBooksAboutContact
HomeBlogPortfolioBooksAboutContact
James Hu·© 2026·All rights reserved
Portfolio·Blog·Books·About·Contact·Privacy

Blog

Writing about system design, ML systems, and building production-grade software.

Featured Series : Architecture, Systems, and Intelligence

Essays on how ML systems behave under real-world constraints.

exploration

Hindsight Is 20/20

A chance remark about cochlear implants turns into an argument that the hard part of learning was never getting the information — it's acquiring the representation that lets you read it.

Jun 24, 2026 · 23 min

commentary

The C++ Renaissance

How AI Collapses and Rewrites the Constraints That Created Modern Languages.

Jan 19, 2026 · 7 min

architecture

I Don't Debug Code — I Debug Agreements

Systems fail at boundaries, not implementations. A practice for writing falsifiable contracts and debugging by auditing agreements instead of tracing execution.

Dec 15, 2025 · 11 min

All Posts
Explore topics:
book (6)
ai (3)
ai-engineering (3)
applied-cs (3)
llm (3)
machine-learning (3)
software-engineering (3)
genomics (2)
01What Production Actually DemandsThe concerns no language teaches you and every system grows into — a companion to the Cross-Cutting Concerns part of my book, Applied Computer Science.Concept · Jun 29, 2026 · 2 minConcept · Jun 29, 2026 · 2 min
02The Repo Is the ProofWhy a concept you can't run is a concept you don't quite believe — a companion to the build-it-yourself approach behind my book, Applied Computer Science.Concept · Jun 29, 2026 · 2 minConcept · Jun 29, 2026 · 2 min
03Learn the Concept, Not the SyntaxWhy "learning a new language" is mostly relearning ideas you already know — a companion to Part I of my book, Applied Computer Science.Concept · Jun 29, 2026 · 2 minConcept · Jun 29, 2026 · 2 min
04Hindsight Is 20/20A chance remark about cochlear implants turns into an argument that the hard part of learning was never getting the information — it's acquiring the representation that lets you read it.Exploration · Jun 24, 2026 · 23 minExploration · Jun 24, 2026 · 23 min
05How to Think About AI AgentsA working mental model for agents — and when not to build one. A companion to Chapter 8 of my book, AI Engineering.Concept · May 29, 2026 · 2 minConcept · May 29, 2026 · 2 min
06RAG Is an Architecture, Not a FeatureWhy retrieval-augmented generation is a system you design, not a library you import — a companion to Chapter 7 of my book, AI Engineering.Concept · May 29, 2026 · 2 minConcept · May 29, 2026 · 2 min
07Building Your First LLM ApplicationThe gap between calling an API and shipping an app — a companion to Chapter 4 of my book, AI Engineering.Concept · May 29, 2026 · 3 minConcept · May 29, 2026 · 3 min
08Neural Networks Got Too Big to Read - Now We Sequence ThemWhy large language models stopped reminding me of software and started reminding me of genomics — and what that says about studying any system too large to read.Exploration · May 28, 2026 · 10 minExploration · May 28, 2026 · 10 min
09The other ninety-eight percentWhat the exome era of genomics reveals about subquadratic attention — and the belief trap that follows a good, necessary compromise.Exploration · May 26, 2026 · 5 minExploration · May 26, 2026 · 5 min
10The C++ RenaissanceHow AI Collapses and Rewrites the Constraints That Created Modern Languages.Commentary · Jan 19, 2026 · 7 minCommentary · Jan 19, 2026 · 7 min
11I Don't Debug Code — I Debug AgreementsSystems fail at boundaries, not implementations. A practice for writing falsifiable contracts and debugging by auditing agreements instead of tracing execution.Architecture · Dec 15, 2025 · 11 minArchitecture · Dec 15, 2025 · 11 min
12Why Reading Backwards Is Hard: Endianness in Computing and Directionality in DNAAn exploration of the surprising parallels between byte order in computer systems and strand directionality in molecular biology, revealing how both domains solve the same fundamental problem.Exploration · Dec 10, 2025 · 5 minExploration · Dec 10, 2025 · 5 min