Freek.dev

We explain why Flare's old AI-generated solutions no longer fit the way developers debug with modern coding agents. The replacement is a better workflow built around the MCP server, CLI, Copy for AI prompts, and agent-friendly docs.

Read more [flareapp.io]

Sebastian explains why Amp became his default coding harness, and why user experience now matters more to him than model churn. He also shares the AGENTS.md tweaks that reduced overzealous verification during day-to-day coding.

Read more [sebastiandedeyne.com]

A practical walkthrough for running Claude Code inside Docker's new sbx sandbox, with a custom PHP setup and extra tools for open source work. Benjamin also explains why sandboxing AI coding agents is the sensible default.

Read more [www.beberlei.de]

Ankur Sethi argues that manually retyping LLM-generated code is a practical way to avoid cognitive debt. It slows you down, but helps you understand the code, catch bad suggestions, and keep a strong mental model of your codebase.

Read more [ankursethi.com]

An essay on how AI-assisted development turns engineers into full-time reviewers. We spend our days evaluating AI-generated diffs instead of writing code, and that shift carries a new cognitive fatigue and may widen the senior/junior gap.

Read more [albertoarena.it]

Mattias walks through moving an AI coding setup off a laptop and onto a private VPS. He covers Tailscale, tmux, a browser-based editor, and the small tweaks that make a remote setup practical from any device.

Read more [ma.ttias.be]

There's a third way to automate Claude Code, and unlike /goal and /loop, it doesn't need your laptop open at all. Anthropic just shipped routines: cloud-hosted automations triggered by a schedule, an API call, or a GitHub event.

Read more [albertoarena.it]

A sharp writeup on how a malicious site could trick Claude into exfiltrating memory-derived personal data through its web browsing flow. It walks through the attack chain, the prompt injection setup, and Anthropic's eventual mitigation.

Read more [www.ayush.digital]

Jarred Sumner explains why the Bun team is moving from Zig to Rust, after years of fighting memory safety issues in a codebase that mixes GC and manual memory management. He also shares how Claude helped make a mechanical, test-suite-driven rewrite realistic without pausing feature work for a year.

Read more [bun.com]

Agentic commerce is a buzzy term, but is still pretty abstract for developers asking what they can actually build today. This post explains MPP through a working Laravel example, and introduces the new square1/laravel-mpp package.

Read more [www.conroyp.com]

Nick Houtman shares how AI is reshaping UX work: clients arrive with better prototypes, common patterns get safer and more generic, and real research still matters most.

Read more [spatie.be]

Matt Pocock shows how his /teach skill turns Claude Code into a personalized teacher that adapts lessons to your goals, level, and progress. It creates structured lessons, resources, quizzes, and a learning record so you can keep building on what you learned.

Read more [www.aihero.dev]

Every Laravel RAG tutorial builds the same ingestion pipeline (chunk, embed, store) and stops the moment the agent answers on screen. None of them check whether retrieval is any good. But retrieval quality is decided at ingestion, before the model runs once, and four decisions there fail with no error, no exception, no failed test:

• Chunking that severs the answer mid-sentence, so answer@1 falls while source hit@1 still looks healthy.
• An HNSW index built with vector_l2_ops while you query with cosine <=>. Postgres silently ignores the index and scans every row. Laravel 13's native whereVectorSimilarTo() hardcodes <=>, so it's easier to hit than ever. Shown with EXPLAIN.
• The embedding dimension baked into the vector(1536) column type, so "shrink it to save storage" is a migration plus a full re-embed that quietly drops retrieval to 47%.
• Ingesting and querying with different models, which turns every distance into noise.

Each bug is real code from a working repo, proven against an eval suite. It's the prequel to my earlier "Evaluating RAG in Laravel" post: build it, prove it, tune it. Every example verified against laravel/ai v0.7.2 and pgvector, with the full repo to clone.

Read more [mujahidabbas.dev]

A practical take on multi-agent setups in Laravel: another agent only earns its place when it needs its own model, tools, or instructions. Good piece on delegation tradeoffs, context handoff, and how to test routing before it becomes an expensive latency tax.

Read more [mujahidabbas.dev]

You can't out-prompt an attacker — to the model, your system instructions and a malicious support ticket are the same text. So stop defending the prompt and lock down the boundaries you actually control: tools scoped to the authenticated user server-side, middleware that screens and logs, output handled as untrusted input, a human in front of anything irreversible, and a fake-free test that fails CI the moment someone drops the auth scope.

Read more [mujahidabbas.dev]

Matthias writes that AI has shifted more of software development from typing to thinking, reviewing, and iterating. Nice reflection on how agentic coding, parallel worktrees, and voice dictation can slow individual features down while still increasing overall output.

Read more [ma.ttias.be]

Your Agent::fake() tests prove your Laravel AI feature runs — not that its output is any good. This evals a real ticket classifier with the AI SDK: a golden dataset for the fields you can check, an LLM-as-judge for the free text you can't, and a regression gate that catches a bad prompt before your customers do.

Read more [mujahidabbas.dev]

An interesting take on how AI changes the build-versus-buy decision for packages. Shared problems can often be generated, while hard, external, or opinionated problems still benefit from well-maintained packages.

Read more [spatie.be]

Josh Comeau shares a thoughtful take on AI, arguing that deep technical skill becomes more valuable, not less, as these tools improve. His point is that strong developers can use AI to amplify their work, while less experienced builders still struggle without solid architectural judgment.

Read more [www.joshwcomeau.com]

Keith Rabois shares a sharp, opinionated take on how AI is reshaping startups, careers, and product teams. The episode is especially interesting for its thoughts on hiring, the future of product management, and building with more speed and intensity.

Read more [podcasts.apple.com]