In the devlog-ist/landing project, we're focused on delivering high-quality content. A crucial part of this is ensuring that AI-generated content meets our standards before it's published.

The Problem: Silent Content Truncation

AI models, particularly when generating longer pieces of content, can sometimes be cut short due to token limits or other constraints.

When publishing content to LinkedIn, large image sizes can lead to 413 Request Entity Too Large errors, especially when dealing with base64 encoded images. This post discusses how we addressed this issue in the devlog-ist/landing project by implementing server-side caching of images.

The Problem: Large Image Payloads

Directly embedding base64 encoded images into API requests,

This post explores the recent updates to cost tracking and budgeting for AI operations within the devlog-ist/landing project.

The Challenge

As AI models become more integrated into our applications, managing and controlling the costs associated with their usage is crucial. Without proper tracking and budgeting, expenses can quickly spiral out of control, especially when dealing with

Introduction

This post explores the implementation of cost tracking and budget limits for AI operations within the devlog-ist/landing project. We'll examine how per-request costs are tracked in EUR, how pricing is stored per model, and how monthly cost limits are enforced per plan.

Cost Tracking in EUR

The system now tracks the cost of each AI operation in EUR. This involves:

This post details the recent enhancements to the content generation process within the devlog-ist/landing project, focusing on the integration and management of LinkedIn prompts for improved content quality and platform-specific tailoring.

The Goal

The primary objective was to enhance the content generation workflow by incorporating LinkedIn-specific prompts, allowing for more targeted and

Working on the devlog-ist/landing project, we're refining the authentication flow. The goal is to streamline user registration and login, particularly for non-developers who might be contributing recommendations. We've adjusted the available OAuth providers to match user roles.

Previously, both GitHub and LinkedIn were presented as default options on the login and registration pages.

OAuth provides a streamlined approach to user authentication, but it's crucial to implement security measures that protect user data and prevent unauthorized access. A common scenario involves social login, where users authenticate via third-party providers like GitHub or LinkedIn.

The Problem: Unregistered Users and Automatic Registration

A potential vulnerability arises when users not yet

Introduction

In our ongoing effort to improve the reliability and traceability of email communications within our application, we've recently implemented transactional email headers for recommendation notifications. This, coupled with some code style enhancements, aims to provide a more robust and maintainable system.

Transactional Email Headers

Transactional emails, such as

Improving the accuracy and scope of technology detection is crucial for generating relevant and informative content. A recent update introduces rule-based technology detection, significantly expanding our ability to identify the technologies involved in code changes. This enhancement allows for more precise tagging and categorization of blog posts, benefiting both content creators and readers.

Our application now supports content negotiation to better serve AI agents and LLMs. We've added the ability to return Markdown instead of HTML when requests include the Accept: text/markdown header.

This enhancement allows AI agents to directly consume the raw Markdown content of our posts, simplifying parsing and improving efficiency.

The Problem

Previously, our application served HTML