Engineering Reads — Week of 2026-05-14 to 2026-05-21#

Week in Review#

This week’s engineering discourse centers heavily on the boundaries of control, specifically how we constrain non-deterministic LLMs into predictable workflows and stop abdicating technical responsibility to our tools. Whether it is defining rigorous feedback loops for coding agents, fighting the structural normalization of memory-safety vulnerabilities, or reclaiming local execution capabilities for frontier AI, the mandate is clear. The mature engineering response to modern complexity is to establish rigorous, observable boundaries rather than surrendering to the path of least resistance.

Engineering Reads — Week of 2026-04-17 to 2026-05-01#

Week in Review#

This week’s reading fundamentally re-evaluates the role of the software engineer in an era where text and code generation are practically free. The dominant debate has shifted from how to generate logic faster to how we deterministically verify it, forcing a transition toward strict mechanical guardrails and “agentic engineering”. Alongside this technical shift, there is a fierce resurgence in confronting the sociopolitical reality of our craft, reminding us that architectural choices—from open-source licenses to structural capability boundaries—never exist in a moral vacuum.

Simon Willison — Week of 2026-04-18 to 2026-05-01#

Highlight of the Week#

The alpha release of llm 0.32a0 marks a foundational architectural pivot for Simon’s ecosystem of CLI tools. By moving away from a simple text-in/text-out abstraction to one that natively models complex message sequences and typed streams, the library is now future-proofed to handle the realities of modern frontier models. This opens the door for seamless integration of server-side tool calls, multi-modal inputs, and reasoning tokens.

Engineering Reads — Week of 2026-05-07 to 2026-05-15#

Week in Review#

This week’s engineering discourse reflects a mature industry grappling with system boundaries and human intent. From constraining unpredictable AI integrations into strictly bounded functional workflows to leveraging organizational psychology to structure open-source compiler architecture, practitioners are aggressively reclaiming control over non-determinism. We are seeing a distinct pushback against buzzword-driven hype in favor of operational stability, rigorous domain modeling, and trusting native web standards over heavyweight abstractions.

Simon Willison — Week of 2026-05-08 to 2026-05-15#

Highlight of the Week#

The standout development this week is Simon’s rapid adaptation to the latest frontier model capabilities, most notably releasing llm 0.32a2 to expose and visualize the new interleaved reasoning tokens of GPT-5 class models directly in the terminal. This perfectly pairs with his hands-on explorations of embedding LLM calls deeply into developer workflows, such as executing prompts via script shebangs and leveraging models to output rich HTML rather than just Markdown.

Engineering Reads — 2026-04-28#

The Big Idea#

The transition of LLMs from individual coding assistants to team-wide engineering tools requires treating prompts as first-class, version-controlled artifacts. We are shifting from ad-hoc interactions with AI to a structured workflow where prompts demand abstraction-first thinking and dictate business alignment.

Deep Reads#

[Structured-Prompt-Driven Development (SPDD)] · Wei Zhang and Jessie Jie Xia · MartinFowler.com While LLM coding assistants have proven valuable for individual developers, scaling their impact across engineering teams requires formalizing how we interact with them. Thoughtworks’ internal IT organization has developed a workflow called Structured-Prompt-Driven Development (SPDD), which treats prompts not as ephemeral chat logs, but as first-class engineering artifacts stored alongside code in version control. By formalizing prompts, teams can better align generated code with actual business requirements. However, this shift demands a change in engineering muscle; developers must index heavily on “abstraction-first” thinking, continuous alignment, and rigorous iterative review rather than relying on the LLM for architectural direction. Practitioners navigating the messy transition from “AI as a toy” to “AI as a predictable team multiplier” should read this to see a concrete, version-controlled approach to prompt management.

Simon Willison — 2026-04-29#

Highlight#

The standout update today is the alpha release of llm 0.32a0, which introduces a major architectural shift to handle the complex realities of modern frontier models. By moving from a simple text-in/text-out abstraction to one based on message sequences and typed streaming parts, Simon is future-proofing the library to seamlessly support reasoning tokens, server-side tool calls, and multi-modal inputs and outputs.

Posts#

[LLM 0.32a0 is a major backwards-compatible refactor] · Source Simon has released an alpha version of his LLM Python library and CLI tool that significantly refactors how models process prompts and responses. Recognizing that modern LLMs possess complex capabilities like reasoning, executing tool calls, and returning images or audio, the original text-in/text-out abstraction was no longer sufficient. The library now models inputs as a sequence of conversational messages and outputs as a stream of typed message parts. Developers can use the new llm.user() and llm.assistant() builder functions to cleanly feed in previous conversation turns without relying on SQLite, while the updated streaming interface elegantly interleaves text, tool execution requests, and reasoning output. For CLI users, the only visible change is a new -R/--no-reasoning flag that suppresses thinking tokens, and Python API users gain a new built-in serialization mechanism to roll their own storage alternatives.

Simon Willison — 2026-05-12#

Highlight#

The standout update today is the alpha release of llm 0.32a2, which adapts to OpenAI’s new endpoints to expose interleaved reasoning across tool calls for GPT-5 class models. It’s a great example of Simon quickly evolving his CLI tools to make the latest LLM reasoning capabilities highly visible and practical for developers.

Posts#

llm 0.32a2 · Source Simon dropped a crucial update to his llm CLI to support the latest reasoning-capable OpenAI models (like the GPT-5 class), which now use a different endpoint rather than /v1/chat/completions. This shift enables interleaved reasoning across tool calls, and the CLI now natively displays these summarized reasoning tokens in a distinct color directly in the terminal. For those who prefer a cleaner output, you can easily suppress the reasoning steps using the new -R or --hide-reasoning flags.

Engineering Reads — 2026-05-15#

The Big Idea#

The maturation of native web standards is eroding the necessity of heavyweight utility frameworks, allowing engineers to reclaim simplicity by lifting framework concepts directly into native implementations. Concurrently, open-source communities are being forced to enact strict moderation boundaries to protect engineering velocity from sprawling ideological debates.

Deep Reads#

Moving away from Tailwind, and learning to structure my CSS · jvns.ca Transitioning away from a framework like Tailwind doesn’t require abandoning its structural lessons; rather, engineers can extract its underlying systems—such as preflight resets, utility classes, and typographic scales—and implement them directly in semantic CSS. The author restructures their plain CSS into conceptual components with unique classes, effectively treating stylesheets like isolated Vue or React components to prevent global cascading failures and keep cognitive overhead low. Instead of relying on Tailwind’s predefined media query utilities (e.g., md:text-xl), the native architecture heavily leverages modern CSS Grid features like auto-fit and minmax() to construct fluid, responsive layouts without arbitrary breakpoints. The primary tradeoff of dropping the framework is losing its built-in guardrails and relying entirely on personal discipline, though combining native CSS @import and nesting capabilities with a minimal esbuild pipeline helps maintain project sanity. Full-stack developers and frontend engineers should read this to understand how modern CSS standards have caught up to utility frameworks, offering the flexibility to write complex layouts that strict utilities fundamentally restrict.

Simon Willison — 2026-05-16#

Highlight#

The standout update today is the release of datasette-llm-limits 0.1a0, which introduces a practical way to manage LLM API costs directly within Datasette. It’s a highly useful piece of infrastructure for anyone building and exposing AI tools, solving the very real problem of managing usage limits for local or hosted LLM integrations.

Posts#

[datasette-llm-limits 0.1a0](https://simonwillison.net/2026/May/15/datasette-llm-limits/#atom-everything) Simon released an alpha version of datasette-llm-limits, a new plugin that works alongside the datasette-llm and datasette-llm-accountant packages. It allows administrators to configure per-user or global spending limits for LLM usage inside of Datasette. This is a crucial addition for safely scaling AI-assisted database workflows by keeping API usage costs strictly under control.