Simon Willison's blog offers a mix of technical tutorials, data analysis projects, and reflections on technology and society. With a focus on Python, SQL, and web development, Simon shares insights into building web applications, working with data, and exploring emerging technologies. Developers can learn about data visualization, API design, and software engineering best practices, gaining inspiration and practical knowledge to advance their careers.

Simon Willison

StrongDM's AI team has implemented a "Software Factory" approach where coding agents write and test software without human code review. Their system uses scenario testing with holdout sets, a Digital Twin Universe that clones third-party services like Okta and Slack for unlimited testing, and aims to spend at least $1,000 per day per engineer on AI tokens. The three-person team builds security software by having agents create API clones from documentation, then runs swarms of simulated testers against the code. They've released Attractor (a spec-only repo for building coding agents) and cxdb (an AI context store). This represents a radical shift where engineers design systems that build code rather than writing it directly.

How StrongDM’s AI team build serious software without even looking at the code

<p>Now I’m keen to implement this technique (with far less budget 😅 )</p>
