⛵NSWE

Blog

Latest posts

Articles on software, teams, and systems.

NSWE

The Session PR: Reviewing How the Code Was Made, Not Just What Changed

aicoding-agentscode-reviewpull-requestsagentic-engineeringsoftware-engineering

Today's pull request is built around a diff. You review what changed, not how it came to be. With coding agents, that model is starting to leak: a clean diff can hide a messy session, skipped constraints, or a plan that was wrong for three iterations before it was right. This post explores a new kind of PR — one where the artifact under review is the agent session itself: the prompts, the tool calls, the rejected paths, the plan, and the human's choices along the way. It compares the idea to existing practices like AI review bots, stacked diffs, build provenance, and "session provenance", and argues that making the session the primary review surface is genuinely new — and probably inevitable.

Set sail

From "Mystery Menus" to Software 3.0: The End of the Middleman

software-3-0aiandrej-karpathyagentic-engineering

The jump from a text-heavy menu to an AI-enhanced visual guide is more than a UX trick, it is a concrete example of Software 3.0. Building on Andrej Karpathy’s Sequoia talk, this post explores the transition from Software 1.0 (explicit code) to Software 2.0 (trained neural networks) to Software 3.0 (LLMs as interpreters). As models increasingly operate directly on user context, many “middleman” apps and interfaces will disappear. The engineer’s value shifts from writing glue code to directing outcomes with judgment, taste, and systems-level understanding.

Set sail

Ownership in the Age of AI-Generated Code

aiengineeringsoftware-engineeringownership

AI-generated outputs from tools like Claude and GitHub Copilot are not independent artifacts but the direct result of how they are guided through prompts, context, and constraints. This means the engineer fully owns both the strengths and flaws of the output. Selective attribution, where success is claimed and failure is blamed on the model, is inconsistent and undermines standards. Effective use of AI requires deliberate input, rigorous review, and full accountability, with the understanding that anything produced and shipped is ultimately the engineer’s responsibility.

Set sail

Open-Source AI in Cybersecurity: Massive Leverage, Real Job Disruption, and What to Do Next

cybersecurityaifuture-of-workopen-source

Open-source AI in cybersecurity is accelerating both offense and defense. Small teams now have capabilities that previously required large budgets, while companies simultaneously rethink security headcount and role design. This post breaks down the opportunity, the disruption risk, and the practical adaptation strategy for engineers and security professionals who want to stay ahead.

Set sail