Without evals, every change is a guess
Why an AI agent needs evals before its next prompt or model change, what makes an agent harder to evaluate than a single LLM call, and the task, run and grade loop every eval is built on.
Posts
Notes and tutorials on Laravel, PHP, electronics and whatever I'm building.
Some posts are meant to be read in order: browse the series.
Who judges the input: screening prompt injection with Laravel Judgment
Putting a separate judge in front of an LLM-based CV screener: Laravel Judgment asks Jev whether a CV contains instructions aimed at the machine, and the application decides what to do with the answer.
When you can't write the rule: Laravel Judgment with Jev and Laya
How to let a model assess free-text refund requests with Laravel Judgment while the application keeps the decision, first on the hosted Jev and then on a self-hosted Laya server.
Content is not an instruction: prompt injection countermeasures in Laravel AI
How a line of hidden text in a CV can steer an LLM-based screener, and the countermeasures from LLM Application Engineering that limit it: separate instructions, checks before and after the call, least privilege and approval gates.
Create a custom validation rule with string alias
In this tutorial I will show how to define a Laravel validation rule providing a custom alias string to use in custom validation classes.
Learning PHP 8.1 Fibers using Tests
PHP 8.1 introduced the new Fiber class (see the proposed RFC here https://wiki.php.net/rfc/fiber), which allows running coroutines inside your script, i.e. two functions executing simultaneously
Sending a test email from a Laravel application
Learn how to send a test email from a Laravel app
What's behind Laravel Encryption/Decryption
My system is safe, it uses encryption. You heard/said this from time to time. Sure, but why and how is it safe? Do you really know this?
Database-oriented Strategy configuration
This article describes a way of implementing Strategy design pattern using a database-based runtime configuration. The approach allows to dynamically (re)configure class behavior without requiring code-based rearrangements.
Refactor long methods to method classes
Classes with a lot of lines of code are generally bad. Too many responsibilities, low cohesion, or both. However, sometimes it is not possible to make functions smaller. Reasons could be various. What can be done in this circumstance?