Freek.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.

How we use class-based Laravel Pennant features, with a kill switch on every flag and a config-driven path to general availability.

Learn how Expressive can improve a Laravel application by keeping Eloquent as the database layer while moving business logic to fully typed objects.

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.

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.

A deep dive into the Laravel request lifecycle, from public/index.php and application bootstrapping to middleware, routing, controllers, responses, and termination.

Our MCP server gives LLMs fast, structured access to Laravel plugin health, compatibility, and vendor reputation, so they recommend maintained, safe choices.

A deep dive into idempotency, from the theory behind safe retries to a practical Laravel implementation using the Laravel Idempotency package.

Learn how to reuse your Laravel MCP tools in the Laravel AI SDK using a simple proxy approach, helping you avoid code duplication and making it easier to build AI agents integrated into your Laravel applications.

Running SQLite on a Laravel Forge site with zero-downtime deployments? Every deploy quietly throws your entire production database away. Here is why it happens and how to fix it by moving the database into the shared storage directory.

A Claude Code skill that helps you design and generate event-sourced domain code for Laravel using spatie/laravel-event-sourcing

Laravel Boost only reads a single core.blade.php per package, so extra sibling files get silently dropped. Here is the minimal pattern for splitting your guidelines into organized partials using a Blade view namespace.

A deep dive into Role-Based Access Control, from the theory behind roles and permissions to a practical, team-aware Laravel implementation without external packages.

A deep dive into how Laravel transforms raw route segments into models, scoped children, enums, and custom bound values before your controller runs.

AI coding agents love to run tests in parallel processes. That's great until multiple processes try to use the same local test database at once. A small file lock can serialize access and stop those runs from stepping on each other.

This article frames AI as a tool to support, not replace, developers, emphasizing the importance of staying in control of how and when it’s used. It encourages a thoughtful approach where developers leverage AI for efficiency while maintaining ownership of decisions and outcomes.