Eleventy, often abbreviated as 11ty, is a static site generator (SSG) designed to simplify the process of building static websites. Developed using JavaScript, it is known for its flexibility and ease of use. 11ty doesn&#39;t impose a specific structure or template language, allowing developers to use their preferred tools and templates.

default

11ty

11ty Framework

Ansible is a powerful automation tool designed for IT tasks such as configuration management, application deployment, and task automation. It uses an agentless architecture, which means it can manage systems without needing additional software installed, making it ideal for web development environments where simplicity and efficiency are key. Its declarative language and playbook structure allow developers to define the desired state of their systems and applications clearly.

Ansible

Ansible Automation Tool

Anthropic Claude is an AI assistant API provided by Anthropic, offering advanced language model capabilities as an alternative to OpenAI&#39;s GPT models. Claude is designed with a focus on safety, helpfulness, and following instructions accurately. The API provides access to Claude models with different capabilities and sizes, suitable for various use cases from simple text generation to complex reasoning tasks. Claude models are known for their long context windows, making them particularly useful for processing large documents and maintaining context in extended conversations. Claude&#39;s API is compatible with similar patterns to OpenAI&#39;s API, making it relatively straightforward to integrate into existing applications. It&#39;s particularly valuable as a backup or alternative provider, providing redundancy and allowing teams to choose the best model for specific tasks. The models are well-suited for applications requiring careful, nuanced responses, content moderation, and tasks where safety and accuracy are priorities. Claude&#39;s focus on being helpful, harmless, and honest makes it a good choice for customer-facing AI applications. Anthropic provides comprehensive documentation and developer tools, and the API integrates well with LangChain and other AI frameworks. Having multiple LLM providers in your toolkit provides flexibility, cost optimization opportunities, and resilience against API outages. With its focus on safety and helpfulness, Claude has become a popular alternative to OpenAI, especially for applications where these qualities are particularly important, making it valuable for teams building production AI applications.

Anthropic Claude

Anthropic Claude API

Apache Superset is an open-source platform for exploring data and building dashboards on top of SQL databases. It queries the source live and keeps no copy of the data itself. It fits reporting on application data that has to stay inside the operator&#39;s infrastructure, where licence costs or vendor lock-in rule out commercial BI tools. We pair it with a read-only database connection and SQL views that flatten application structures into report-friendly tables. Dashboards can be embedded into an application with guest tokens and row-level security, or used directly in Superset.

Apache Superset

Apache Unomi provides a server which stores and queries customer data. With that data, users are placed into multiple segments, based on their activity on the site. This allows content to be targeted to specific segments, all while respecting visitor privacy rules and keeping all visitor data within the site&#39;s infrastructure rather than tracking users across the rest of the web.

Apache Unomi

Astro is a modern static site generator that focuses on content-driven websites with minimal JavaScript. It allows developers to build fast websites by shipping zero JavaScript by default, only adding interactivity where needed through &quot;islands architecture.&quot; The framework is framework-agnostic, allowing developers to use components from React, Vue.js, Svelte, Preact, or plain HTML/CSS/JavaScript in the same project. This flexibility means teams can leverage their existing component libraries and knowledge while building static sites. Astro&#39;s key innovation is its islands architecture, where interactive components are isolated and hydrated independently. This means the rest of the page remains static HTML, resulting in extremely fast page loads and excellent Core Web Vitals scores. The tool provides excellent developer experience with features like Markdown and MDX support, image optimization, automatic code splitting, and built-in support for TypeScript. It integrates well with popular headless CMSs and can be deployed to various hosting platforms. Astro is particularly well-suited for content-heavy websites, blogs, documentation sites, and marketing pages where performance is critical. It competes with tools like 11ty but offers a more modern developer experience with component support and better integration with modern tooling. With its focus on performance, developer experience, and flexibility, Astro has gained significant traction in the static site generation space and is worth evaluating as an alternative or complement to 11ty for projects where component-based development and modern tooling are priorities.

Astro

Astro Static Site Generator

Atelier is an open-source Drupal distribution for AI-assisted page building, built on FlowDrop . Editors connect their own keys for commercial providers, OpenAI-compatible endpoints or local models, and assign models to roles. It fits standalone landing pages and campaign sites where an editor should get from raw text to a finished page quickly. As a distribution, it brings its own theme, components and configuration, which makes it a separate application rather than an add-on to an existing site.

Atelier

Atelier AI-Assisted Drupal CMS

BackstopJS is an open-source tool that allows developers to catch visual changes in their web applications by taking screenshots of web pages and comparing them against a baseline. It is particularly useful for ensuring that UI updates do not inadvertently alter the appearance of a site. With customizable configurations and integration capabilities, BackstopJS enhances the testing workflow and improves collaboration among development teams.

BackstopJS

BEM is a CSS naming convention that promotes modularity and reusability in web development. By breaking down the design into &#39;blocks&#39;, &#39;elements&#39;, and &#39;modifiers&#39;, developers can create a clear relationship between HTML and CSS, making it easier to understand and modify styles without affecting unrelated components. This methodology not only streamlines collaboration among teams but also improves the scalability of web applications.

BEM (Block Element Modifier)

Cinny is an open-source client for Matrix , available in the browser and as a desktop app. It focuses on a clean, simple interface for rooms, spaces and direct messages. As a web client it can be hosted next to your own Matrix server, so a team can chat in a familiar interface while the data stays on infrastructure it controls.

Cinny

Cinny Matrix Client

Circle Dot is a comprehensive Drupal installation that includes multiple components styled in the Factorial design. It serves as a foundation for new projects, designed to be adapted and further developed according to the project&#39;s specific needs. There were several reasons for the creation of Circle Dot at Factorial. Setting up a project often takes considerable time, especially when working on Drupal projects where the same tasks are frequently repeated: installing Drupal with various modules, setting up a component library, configuring Continuous Integration and deployments, writing standard components, and more. Estimating and pricing projects is also challenging, as stakeholders need to know the costs, but these estimations are difficult to align with reality. Factorial maintains high standards for code quality, automated testing, and overall product quality, which adds significant value for customers but makes smaller projects less suitable for our approach. Onboarding new developers is another challenge, as each company has its own methods, and new developers need to understand Factorial&#39;s way of working. Furthermore, customers prefer to see tangible examples of what they will receive, as hands-on demonstrations are more convincing than theoretical discussions. Given this background, Circle Dot aims to achieve several goals: First, it seeks to increase efficiency while maintaining high quality, allowing Factorial to offer projects at lower costs or provide more value for the same price, ultimately improving our financial position. Second, it enables production teams to concentrate on the complex aspects of a project by providing a set of proven standards, thereby avoiding the need to constantly reinvent the wheel. Third, it simplifies the process of estimating and pricing projects, as a substantial portion of the costs can be covered by a well-developed and priced base. Fourth, it serves as a go-to resource for new developers, offering well-proven examples of Factorial&#39;s practices. Finally, it acts as a showcase of our work, making it easier to convince potential clients of our expertise by demonstrating a standard setup.

Circle Dot

Circle Dot, Factorial's Drupal CMS base install

CSS Nesting brings a widely used feature from pre-processors natively into the language.

CSS Nesting

Cynefin is a descion making framework, it helps to make sense of different situations to choose an appropriate response. The first step in applying the framework is to become aware of the situation you&#39;re in and than choose appropriate actions. The framework is divided into five domains: clear, complicated, complex, chaotic and disorder

Cynefin

Cynefin Decision Making Framework

Cypress is an open-source testing framework that allows developers to create robust and reliable tests for their web applications. It operates directly in the browser, providing real-time feedback and an interactive testing experience. With features like time travel, automatic waiting, and easy debugging, Cypress simplifies the process of writing tests and helps ensure high-quality user experiences.

Cypress

Cypress Testing Framework

D3.js is a JavaScript libary which we use to create bespoke data visualizations. Developers bind data to DOM elements and apply data-driven transformations to create visually appealing and informative graphics.

D3.js

DDEV is an open-source development environment designed to simplify the setup and management of local development environments for PHP-based web applications, particularly those using content management systems like Drupal, WordPress, and TYPO3. It leverages containerization technologies like Docker to provide a consistent and replicable environment, ensuring that developers can work efficiently across different systems and configurations. By automating many aspects of the development workflow, including environment setup, configuration, and dependency management, DDEV helps developers focus more on coding and less on troubleshooting environment issues.

DDEV

DDEV Development Tool

Drupal is a high-perfomance content management system (CMS) and framework for developing content-heavy websites, self-management portals and generally systems that need to integrate with services inside or outside our clients enterprise IT ecosystem. Drupal is extremely modular and versatile, stable and flexible in use and unites the world’s largest open source community behind Linux.

changed

Changed

Drupal

Drupal CMS

esbuild is a fast JavaScript bundler and minifier, written in Go, that transforms and bundles JavaScript, TypeScript, and other web assets. It supports modern JavaScript features, tree shaking, code splitting, and handles CSS and JSON. Known for its speed, esbuild is used to significantly reduce build times, enhancing developer productivity.

esbuild

ESLint is a static code analysis tool for identifying problematic patterns in JavaScript code. It allows developers to enforce coding styles and catch errors early in the development process, improving maintainability and reducing bugs. With numerous plugins and configuration options, ESLint can be tailored to fit various coding standards and project requirements, making it a crucial asset for any web development team.

ESlint

Fabric is a Python library designed to simplify the process of deploying applications to remote servers. It provides a high-level interface to SSH, allowing developers to automate the deployment of code, manage server tasks, and streamline the release process with ease.

Fabric

Fabric Deployment Tool

FastAPI is designed to create web APIs quickly and efficiently, leveraging Python&#39;s type hints to provide automatic validation, serialization, and documentation generation. It is built on top of Starlette for the web parts and Pydantic for the data parts, giving it exceptional performance and ease of use, making it a popular choice for developing RESTful services and asynchronous programming.

FastAPI

FastAPI Framework

Figma is a a web-based tool for collaboratively building interface design and prototyping by designers. Figma&#39;s Dev Mode in turn is a developer-focused interface for developers inspecting and navigating designs. It makes it easier for developers to get information about the design out and use it for implementing UI components. Dev Mode reduces the needed communication between designers and developers. Both disciplines may add annotations and discuss directly on design. As the design evolves, developers can keep track of changes and compare versions of the designs.

Figma Dev Mode

Filament leverages the power of Laravel to provide developers with a robust set of tools for creating dynamic and interactive admin panels. It allows for rapid development with its elegant design and intuitive API, making it an excellent choice for developers looking to streamline their workflow and enhance user experience in backend management.

filament

Filament Admin Panel for Laravel

FlowDrop is an open-source workflow orchestration engine with a visual editor. A workflow is a graph of typed nodes with explicit edges, and every run can be traced back to the steps it took. It is useful where an AI feature is a pipeline rather than a single model call: load context, build a prompt, call a model, validate and map the answer. Narrow, deterministic steps keep language models on a short leash. FlowDrop runs inside Drupal, connected to its content, users and permissions. A standalone engine written in Rust is in development and currently in a closed beta without public access. A published specification describes the behaviour both implementations follow. The Drupal distribution Atelier is built on it.

FlowDrop

FlowDrop Workflow Orchestration

Fluent is a powerful and flexible localization system for natural-sounding translations. Its custom syntax supports basic key/value string translations, variables, value formatting, all the way up to complex conditions for plurals, genders, or cases. Official Fluent implementations are available for JavaScript , Rust , and Python .

Fluent

Go, also known as Golang, is a statically typed, compiled programming language created by Google. It emphasizes simplicity, efficiency, and concurrency, making it well-suited for building scalable and robust software systems. With its clean syntax, built-in support for concurrency, and extensive standard library, Go is popular for web development, cloud computing, and system programming tasks.

Go Programming Language

Google Firebase is a development platform that has little to desire. It offers a wide range of services out of the box. It is a great choice to build a web app that needs good persistence quickly.

Google Firebase

Grafana is a leading open-source analytics and monitoring solution that allows developers to visualize data from a variety of sources in customizable dashboards. It supports multiple data sources such as Prometheus, InfluxDB, and Elasticsearch, enabling teams to gain insights into application performance, infrastructure, and operational metrics. With its rich features, including alerting, reporting, and sharing capabilities, Grafana empowers developers to make informed decisions based on real-time data.

Grafana

Grafana Monitoring Platform

Helm is the package manager for Kubernetes. It bundles an application&#39;s manifests into a chart with templated values, which is installed, upgraded and rolled back as one release. It is the usual way to run third-party software such as search engines or analytics tools on a cluster, and to deploy your own application to several environments with different values.

Helm

Histoire is a modern tool based on Vite, which can be used to build a company&#39;s design system or a component Library.

Histoire

Histoire Tool

Hugging Face is a platform that democratizes access to machine learning by providing a vast library of pre-trained models, datasets, and tools. It&#39;s particularly known for its Transformers library, which provides easy access to state-of-the-art natural language processing models. The platform offers thousands of open-source models that can be used as alternatives to proprietary APIs, giving developers more control over their AI applications. Models can be run locally or deployed to Hugging Face&#39;s infrastructure, providing flexibility in how AI capabilities are integrated. Hugging Face Transformers library supports Python and provides a consistent API for working with different model architectures. It integrates well with PyTorch and TensorFlow, and includes tools for fine-tuning models on custom datasets. The platform is valuable for teams that want to use open-source models, need to run models locally for privacy or cost reasons, or want to fine-tune models for specific use cases. It&#39;s particularly useful for applications where data privacy is important, as models can be run entirely on-premises. With its extensive model library, active community, and commitment to open-source AI, Hugging Face has become the go-to platform for accessing and deploying open-source AI models, making it an essential tool for teams building AI applications with open-source alternatives.

Hugging Face

Hugging Face Platform

JSON:API is a specification for building HTTP APIs in JSON. One of its goals is to stop bikeshedding and allow developers to focus on building the actual application. When a server follows the JSON:API specification, a client can expect the same structure in the response for every endpoint. This makes it easier to build clients that can consume any JSON:API-compliant API.

JSON:API

Kubernetes, often abbreviated as K8s, is a powerful system for managing containerized applications across a cluster of machines. It provides tools for deploying applications, scaling them as needed, managing changes to existing containerized applications, and helps facilitate the overall management of containerized environments, making it an essential tool for modern web development and DevOps practices.

Kubernetes

LangChain is a framework designed to simplify the development of applications powered by large language models (LLMs). It provides a comprehensive set of tools and abstractions for building AI applications, including prompt management, chain composition, and integration with various LLM providers. The framework supports multiple LLM providers including OpenAI, Anthropic, Hugging Face, and local models. It offers components for common AI application patterns like retrieval-augmented generation (RAG), agents, and memory management. LangChain is particularly valuable for building production AI applications because it handles complex orchestration, manages context windows, and provides patterns for common use cases. It integrates well with Python web frameworks like FastAPI, making it ideal for building AI-powered APIs and services. The framework includes LangSmith for monitoring and debugging LLM applications, and LangServe for deploying LangChain applications as APIs. This makes it a complete solution for developing, testing, and deploying AI applications. With its focus on developer experience and production readiness, LangChain has become the standard framework for building LLM applications, especially for teams integrating AI capabilities into existing web applications.

LangChain

LangChain Framework

Ignition is a replacement error page for PHP. It is part of Laravel, but can be installed in any PHP project, and a Drupal contrib module provides integration. It shows a backtrace with an excerpt of the source code, parameter values, and the details from the current request.

Laravel Ignition

Laravel Nova is a beautifully designed administration panel for Laravel applications that allows developers to quickly create a powerful, customizable backend interface. It offers features like resource management, filters, lenses, and metrics, enabling developers to efficiently manage their application&#39;s data while maintaining a clean and intuitive user experience.

Laravel Nova

Laravel is an open-source PHP framework that eases the development of common tasks used in most web projects and brings the flexibility to build solutions for innovative business needs on any kind of web application. In addition, Laravel has an elegant syntax that is enjoyable to work with and lots of supporting training materials which speed up the learning process and the development itself. Laravel is also the most popular PHP framework and has a large community of developers.

Laravel

Laravel PHP Framework

LlamaIndex is a framework specifically designed for building retrieval-augmented generation (RAG) applications. It provides tools for ingesting, indexing, and querying data from various sources, then using that data to enhance LLM responses with relevant context. The framework supports a wide range of data sources including documents, databases, APIs, and vector stores. It handles the complexity of data ingestion, chunking, embedding generation, and retrieval, making it easier to build applications that combine LLMs with external knowledge. LlamaIndex provides abstractions for common RAG patterns like document Q&amp;A, semantic search, and data-augmented chatbots. It integrates with various LLM providers and vector databases, giving developers flexibility in their technology choices. The framework is particularly valuable for applications that need to provide LLMs with access to private or domain-specific data. It&#39;s commonly used for building intelligent document search systems, knowledge bases, and chatbots that can answer questions about specific content. LlamaIndex works well with LangChain and can be integrated into web applications built with FastAPI or other Python frameworks. It provides both high-level abstractions for quick prototyping and low-level control for custom implementations. With its focus on RAG applications and data integration, LlamaIndex has become a popular choice for teams building AI applications that need to combine LLMs with external data sources, making it essential for knowledge-intensive AI applications.

LlamaIndex

LlamaIndex RAG Framework

LTI is a standard developed by IMS Global that allows different educational tools and platforms to communicate with each other securely and efficiently. It simplifies the process of integrating third-party tools into learning management systems, enabling educators to enrich their courses with a variety of resources and applications without the need for complex development or configuration.

Learning Tools Interoperability (LTI)

MariaDB is a community-developed fork of MySQL that provides a robust and flexible database solution with advanced features such as improved query optimization, dynamic columns, and various storage engines. It is widely used in web applications due to its compatibility with MySQL and its commitment to maintaining an open-source ethos, ensuring developers have access to powerful database capabilities without vendor lock-in.

MariaDB

MariaDB Database

Matomo is an open-source web analytics platform that can run on your own infrastructure. It records page views and custom events sent from the browser or from the server. It suits applications where usage data must not leave the operator&#39;s infrastructure. Server-side events let you measure behaviour without client-side tracking, for example where consent management is to be avoided.

Matomo

Matomo Analytics

Matrix is an open-source protocol that facilitates real-time communication and collaboration over the internet. It allows users to communicate securely and seamlessly through various clients, supporting features like text, voice, and video calls, while ensuring data ownership and privacy. Its decentralized nature means that users can host their own servers, providing greater control and flexibility in how communication occurs.

Matrix

Matrix Protocol

Mattermost is a flexible, self-hosted chat platform designed for team collaboration, offering features such as threaded conversations, integrations with various tools, and strong security measures. It is particularly beneficial for organizations seeking to maintain control over their data while enabling efficient real-time communication among team members.

Mattermost

Mattermost Messaging Platform

Mautic is an open source marketing automation tool. It was created with the idea in mind to bring the power of marketing automation to literally everyone by drastically reducing the cost and effort it takes to setup such a system. It provides its users with all means necessary to build powerful newsletter campaigns, landing pages and forms. On top of that organizations are able to track a campaign&#39;s performance using feature rich reports. Also customers can be organized in segments by customizable rule sets for advanced personalization of campaigns. In conjunction with Drupal and Apache Unomi, Mautic is the missing piece to operate a Drupal website as a Digital Experience Platform (DXP) where Apache Unomi puts users into segments by their browsing behaviour, Drupal delivering personalized content based on these segments and Mautic providing the off site marketing tools to interact with customers after they have visited the website.

Mautic

Mautic Marketing Automation

mise is a polyglot tool version manager. It replaces tools like asdf, nvm, pyenv, rbenv, etc. mise allows you to switch sets of env vars in different project directories. It can replace direnv. mise is a task runner that can replace make, or npm scripts.

mise-en-place

miyagi is an open-source tool for developing and documenting UI components, created at Factorial. It renders components written in template engines such as Twig, together with mock data, variations and documentation. It suits server-rendered projects, such as Drupal themes, where components live in templates rather than in a JavaScript framework. Designers, developers and QA get a component library they can browse, deployed alongside the application.

miyagi

miyagi Component Library

MCP is an open protocol for connecting AI assistants such as Claude to tools and data. An application runs an MCP server that describes its tools; any MCP client can then call them. It turns an application into something you can work with in plain language, without a separate integration for every assistant. Many vendors, Atlassian among them, ship their own MCP servers.

Model Context Protocol (MCP)

A monolithic repository (monorepo) is a strategy where the source code of multiple software components is kept together inside a single repository.

Monorepo

Monorepo Strategy

Mock Service Worker (MSW) is an API mocking library for JavaScript. It intercepts requests at the network level, through a Service Worker in the browser and through interceptors in Node.js, and answers them with handlers you define. Because the application keeps making real HTTP requests, the same mocks serve local development, tests and demos without changes to application code. Combined with an agreed OpenAPI contract and generated test data, for example from Faker, frontend work can start before the backend endpoint exists.

Mock Service Worker (MSW)

Observable/plot is a data visualisation library written in JavaScript. It is built on D3 , which is more of a low-level nature. Observable provides heavily configurable chart presets, while D3 provides building blocks for charts. There is also Observable/framework, which acts more like a Static Site Generator for chart-heavy websites, which we didn&#39;t use though, so the following refers to Observable/plot only. Both libraries are built on similar principles, which makes it easy to learn one, if you know the other one already. With observable you get a chart up and running in very low time, in return you are limited to the available chart types (but there are a lot of them) and their respective options for fine-tuning (which are plenty, though).

Observable

Data Visualization with Observable JS

Ollama is a tool that enables running large language models locally on your own hardware. It provides a simple way to download, run, and manage LLMs without requiring cloud APIs, making it ideal for applications where data privacy, cost control, or offline operation are priorities. The tool supports a wide range of open-source models including Llama, Mistral, CodeLlama, and many others. It provides a REST API that&#39;s compatible with OpenAI&#39;s API format, making it easy to swap between local and cloud-based models in applications. Ollama is particularly valuable for development and testing, as it allows developers to experiment with LLMs without API costs. It&#39;s also useful for production applications where data must remain on-premises, or where API costs would be prohibitive for high-volume use cases. The tool runs efficiently on modern hardware, including Apple Silicon Macs, making it accessible for local development. It can be deployed in Docker containers for production use, and integrates well with existing applications through its API. With its focus on simplicity and local execution, Ollama has become a popular choice for teams that want to leverage LLM capabilities while maintaining control over their infrastructure and data, making it an important tool for privacy-conscious AI applications.

Ollama

Ollama Local LLM Runtime

OpenAI API provides programmatic access to OpenAI&#39;s large language models, including GPT-4, GPT-3.5, and specialized models for embeddings, image generation, and audio transcription. It&#39;s the most widely adopted API for integrating AI capabilities into web applications. The API offers a simple, consistent interface for interacting with language models, making it easy to add features like chatbots, content generation, summarization, translation, and code generation to applications. It supports both synchronous and asynchronous requests, with streaming capabilities for real-time responses. OpenAI&#39;s models are particularly well-suited for production applications due to their reliability, performance, and extensive documentation. The API includes features like function calling, which allows models to interact with external tools and APIs, enabling more sophisticated AI applications. The API integrates seamlessly with web frameworks like FastAPI and Laravel, and works well with LangChain for building complex AI applications. It&#39;s commonly used for building AI-powered features in content management systems, customer support systems, and data analysis tools. With its comprehensive documentation, active community, and continuous model improvements, the OpenAI API has become the de facto standard for integrating LLM capabilities into web applications, making it essential for teams building AI-powered features.

OpenAI API

OrbStack is a macOS native application and alternative to Docker Desktop. It is a lightweight, fast, and easy-to-use container runtime that is built on top of Apple&#39;s Hypervisor.framework. It is designed to be a simple and efficient way to run containers on macOS. It is a great tool for developers who want to run containers on their local machine without the overhead of a full virtual machine. OrbStack is free and open source software, and it is actively developed and maintained by the OrbStack team.

Orb Stack

OrbStack Tool

Oxc (the JavaScript Oxidation Compiler) is a collection of JavaScript and TypeScript tools written in Rust. Its linter oxlint and its formatter oxfmt are meant to replace ESLint and Prettier. The main reason to use it is shorter lint and format runs, locally and in CI, with fewer dependencies than a typical ESLint plugin setup. Before switching, check which ESLint plugins and rules a project depends on, since not every plugin has an equivalent yet.

Oxc (oxlint and oxfmt)

Pandas offers versatile data structures, such as DataFrames and Series, with robust functionality that makes it the go-to tool for data cleaning, transformation, and exploration. It empowers users to quickly and effectively gain insights from their data. Its widespread use in data science comes from its ability to perform complex data operations with ease, and its seamless integration with libraries for AI, machine learning, and data visualization. Pandas is our preferred tool for data analysis, fitting seamlessly into our stack to power AI-driven features and deliver dynamic dashboards.

Pandas

Pandas Data Analysis Library

Phabalicious is using configuration stored in a special file in the root of your project to run tasks in a shell. This shell can be provided by a Docker Container, a SSH connection, a Kubernetes pod or a local shell. This means, you can store all your DevOps-scripts in a single file and apply it to a list of configurations. Phabalicious tries to abstract away the inner workings of a configuration and give the user a handful useful commands to run common tasks. Phabalicious integrates nicely with existing solutions like for continuous integration or docker-based setups or diverse hosting environments like Acquia SiteFactory, Lagoon, platform.sh, Kubernetes or other complicated custom IT infrastructures.

Phabalicous

PHP is a programming language designed to deliver webpages. It&#39;s by far the most popular language, powering about 80% of the web, and our language of choice for most web projects because Drupal, Laravel, Symfony, WordPress and other major frameworks are all written in it.

PHP Web Development

Pinecone is a managed vector database service designed specifically for AI applications that require similarity search. It&#39;s optimized for storing and querying high-dimensional vectors, making it ideal for retrieval-augmented generation (RAG) applications, semantic search, and recommendation systems. Vector databases are essential for RAG applications, where you need to find relevant context from a knowledge base to provide to an LLM. Pinecone handles the complexity of vector indexing and similarity search, providing a simple API for storing embeddings and finding similar vectors. The service is fully managed, meaning it handles scaling, indexing, and infrastructure management automatically. It provides low-latency queries even with millions of vectors, making it suitable for production applications. Pinecone integrates seamlessly with LangChain and other AI frameworks. Pinecone is particularly valuable for applications that need to provide LLMs with context from large knowledge bases, such as document Q&amp;A systems, chatbots with domain knowledge, and personalized content recommendations. It&#39;s commonly used alongside OpenAI embeddings to build RAG applications. With its focus on performance, ease of use, and production readiness, Pinecone has become a popular choice for vector storage in AI applications, especially for teams building RAG systems that need reliable, scalable vector search capabilities.

Pinecone

Pinecone Vector Database

Platform engineering is a multi disciplinary approach, informed by DevOps principles, that focuses on improving productivity, by providing a platform that allows development teams to focus on developing and delivering new features fast. For that a platform provides and bundles all the necessary tools and services that are needed to develop, test, deploy and operate software. This includes services like databases, storage, messaging, monitoring, logging, security and compliance services.

Platform Engineering

Playwright is an open-source automation framework developed by Microsoft that allows developers to write reliable tests for web sites and web applications. It supports multiple browsers including Chrome, Firefox, and Safari, and enables cross-browser testing with a unified API. With end-to-end testing features like auto-waiting, capturing screenshots, Playwright simplifies the testing process and enhances developer productivity.

Playwright

Playwright Testing Framework

PostgreSQL is a powerful, open-source object-relational database system with over 35 years of active development. It has earned a strong reputation for reliability, feature robustness, and performance, making it a preferred choice for many modern applications. The database offers advanced features including full ACID compliance, foreign keys, triggers, views, materialized views, and stored procedures. It has excellent support for JSON and JSONB data types, making it suitable for both relational and document-style data storage. PostgreSQL&#39;s full-text search capabilities are built-in and highly performant. PostgreSQL is increasingly preferred over MySQL/MariaDB for new projects due to its advanced features, better performance in many scenarios, and more sophisticated query optimizer. It&#39;s particularly well-suited for applications requiring complex queries, data integrity, and scalability. The database integrates well with modern development stacks, including Laravel, which has excellent PostgreSQL support through Eloquent ORM. It&#39;s commonly used in headless CMS architectures and microservices where its JSON capabilities and advanced features provide flexibility. PostgreSQL&#39;s extensibility allows developers to add custom functions, data types, and even programming languages. It has a vibrant ecosystem of extensions that add capabilities like geographic data processing, full-text search enhancements, and more. With its focus on standards compliance and advanced features, PostgreSQL has become the database of choice for many enterprise applications and modern web development projects.

PostgreSQL

PostgreSQL Database

Prometheus is an open-source monitoring system and time series database that collects metrics from configured targets at specified intervals, evaluating rule expressions, and displaying the results. It is designed for reliability and scalability, making it an ideal choice for modern microservices architectures. With its multidimensional data model and powerful query language (PromQL), developers can easily monitor application performance and system health in real-time. Prometheus provides a robust ecosystem of features including service discovery, powerful alerting capabilities, and a variety of integrations with other tools and platforms. Its data model allows for flexible querying and aggregation of time series data, which is crucial for diagnosing issues and optimizing resource usage in dynamic environments. The thriving community around Prometheus ensures continuous improvement and support, making it a staple for DevOps and site reliability engineering (SRE) teams.

Prometheus

Prometheus Monitoring System

Promptfoo is a tool for prompt testing and evaluating LLM applications. It helps create effective prompts and models by providing benchmarks specific to your needs, speeds up the evaluation process with caching and concurrency, and allows automatic scoring of outputs based on defined metrics.

promptfoo

Python is a versatile programming language widely used in data science for its simplicity and rich ecosystem of libraries. Tools like Pandas, NumPy, and SciPy enable efficient data manipulation and analysis, while Matplotlib and Seaborn simplify data visualization. Python&#39;s seamless integration with machine learning frameworks like TensorFlow and scikit-learn makes it ideal for AI and data-driven applications. Additionally, FastAPI allows quick creation of scalable APIs, making Python a powerful tool for everything from data exploration to deploying real-time AI solutions, serving custom software needs effectively.

Python

Python Programming Language

What is a queue? A queue, is a data structure for the linear collection of elements that follows the First-In, First-Out (FIFO) principle. Queues are commonly used in scenarios where order matters. What is a work queue? A work queue on the other hand, extends the basic concept. With work queues you have the work queue itself, producers that add jobs to it, and consumers (workers) which retrieve jobs from the work queue and process them. Why use work queues? This allows to asynchronously process long-running tasks that would otherwise block a system. A good example is when a video is transcoded whenever a new video asset is uploaded. As workers only take on a job whenever they are able to do, the work queue ensures we don&#39;t overwhelm the computing systems the workers run on. On the other hand workers can be scaled to a number that would utilize the system properly. If you have a workload that can be split up, you&#39;re architecture will benefit from parallel processing. Workload will be distributed to all workers consuming from a queue. Outside distribution, work queues help in decoupling systems. A producer doesn&#39;t need to know how and by whom a job is going to be processed. You can use a different language or framework for the consumer. As a downside work queues don&#39;t provide a good mechanism for feedback. Work queues usually don&#39;t provide means to provide feedback to the producer, i.e. &quot;video transcoded&quot;. RabbitMQ RabbitMQ positions itself more generally as a message broker, through which it is possible among other things to provide work queues. Work queues play a major role in the messaging technology space.

RabbitMQ

React is a declarative, efficient, and flexible JavaScript library for building user interfaces. Created by Facebook (Meta), it has become the most widely used frontend framework in the industry, powering applications for companies of all sizes. React uses a component-based architecture where UIs are built from reusable components. It employs a virtual DOM to efficiently update the user interface, only re-rendering components when their state or props change. This approach results in high performance even in complex applications. The library has a massive ecosystem of libraries, tools, and community resources. It integrates well with modern build tools, state management libraries, routing solutions, and testing frameworks. React&#39;s popularity means there&#39;s extensive documentation, tutorials, and community support available. React&#39;s JSX syntax allows developers to write HTML-like code within JavaScript, making component code more readable and maintainable. The library promotes unidirectional data flow and encourages developers to think about applications in terms of components and their relationships. While Factorial primarily uses Vue.js and Svelte, React is worth having in the radar because many clients may request React-based solutions, and it&#39;s important to understand the most widely adopted framework in the industry. React&#39;s ecosystem and job market presence make it valuable knowledge for any web development team. The library continues to evolve with features like React Server Components, concurrent rendering, and improved developer experience, maintaining its position as a leading frontend technology.

React

React JavaScript Framework

Redis is an open-source, in-memory data structure store that serves multiple purposes in modern web applications. It&#39;s commonly used as a caching layer to dramatically improve application performance by storing frequently accessed data in memory, reducing database load and response times. Beyond caching, Redis supports various data structures including strings, hashes, lists, sets, sorted sets, and streams. This versatility makes it suitable for use cases like session storage, real-time analytics, message queuing, rate limiting, and leaderboards. Redis is particularly valuable in Laravel applications, where it&#39;s commonly used for session storage, queue management, and caching. It integrates seamlessly with Laravel&#39;s cache and queue systems, providing a fast and reliable backend for these features. The database offers persistence options, allowing data to be saved to disk for durability. It supports replication, high availability through Redis Sentinel, and horizontal scaling through Redis Cluster, making it suitable for production environments of any scale. Redis&#39;s performance characteristics make it ideal for applications requiring low-latency data access. Its simple API and excellent documentation have made it one of the most popular NoSQL databases, with widespread adoption across the industry.

Redis

Redis In-Memory Data Store

Rust, a robust systems programming language developed by Mozilla, is recognized for its emphasis on performance, reliability, and safety. This language is defined by three key features. First, its ownership system enforces strict rules on memory management, preventing common programming errors. Second, it offers zero-cost abstractions, allowing developers to use high-level language constructs without compromising performance. Lastly, Rust&#39;s strong type system enhances code safety and prevents type-related errors, contributing to a reliable and secure development environment.

Rust

Rust Programming Language

scikit-learn is a widely-used machine learning library for Python that provides simple and efficient tools for data mining and data analysis. It&#39;s built on NumPy, SciPy, and matplotlib, and provides a consistent API for a wide range of machine learning algorithms. The library includes tools for classification, regression, clustering, dimensionality reduction, model selection, and preprocessing. It&#39;s designed to be easy to use while providing powerful capabilities for both beginners and experienced data scientists. scikit-learn is particularly valuable for building traditional machine learning models, feature engineering, and data preprocessing. It integrates seamlessly with Pandas (which we already use) and other Python data science tools, making it a natural fit for our Python-based AI projects. The library is well-documented, has a large community, and is actively maintained. It&#39;s commonly used for building recommendation systems, classification models, and predictive analytics features in web applications. While deep learning frameworks like TensorFlow and PyTorch are better for neural networks, scikit-learn excels at traditional machine learning tasks and is often the right tool for many practical AI applications. It&#39;s particularly useful for projects that need interpretable models, quick prototyping, or don&#39;t require the complexity of deep learning. With its focus on simplicity, performance, and practical machine learning, scikit-learn is an essential tool for teams building AI features that don&#39;t require deep learning, making it a valuable addition to our AI toolkit.

scikit-learn

scikit-learn Machine Learning Library

Sentry provides developers with comprehensive insights into application errors and performance bottlenecks, allowing for proactive issue resolution. With support for multiple programming languages and frameworks, Sentry integrates seamlessly into development workflows, offering detailed error reports, user context, and performance metrics, thus improving the overall user experience.

Sentry

Sentry Error Tracking

Shared First is an alternative to the Mobile First approach that has been conceived from our practice. In principle, the mobile-first approach can be differentiated between the technical implementation of a mobile-first approach and a design and user-centered mobile-first approach. In the design and user-centered mobile-first approach, page layout and interaction patterns are primarily designed with the mobile usability of the website in mind. Placing important and frequently searched elements in the visible area of the website is the first key, as users can find the content they want more quickly and do not have to scroll through the website or menus for a long time. A central search can also increase user satisfaction because it can provide a direct shortcut to the desired topic. A flat menu hierarchy and directly visible links to topics even at deeper levels offer comparable benefits. Avoiding horizontal scrollers and sliders makes it easier to capture the offer. In addition to design aspects, attention should be paid to short texts when creating content, especially on landing pages, in order to increase the usability of the mobile site. Thanks to faster loading times, mobile users are not held up by long waiting times and the mobile data volume is conserved through reduced data volumes. From a design perspective, the use of large, easy-to-read fonts and Providing sufficiently large touch areas for good accessibility is essential. In addition, it is important to ensure that there is sufficient contrast, as sunlight or reflections, for example, can impair the readability of the content, especially in a mobile context. Easily accessible contact options help particularly restricted users to receive help or information quickly and easily if necessary. The technical-conceptual mobile-first approach was the standard in front-end development for many years. There were technical reasons for this: Some browsers at the time did not yet support the necessary technology (so-called media queries), so they wanted to ensure an appropriate display for the users of these browsers. The disadvantage of this approach, however, is that it is relatively error-prone - visual adjustments can easily have unwanted side effects. Since all browsers now support media queries, mobile-first development is no longer absolutely necessary. At Factorial we therefore use an approach that we call shared-first. This has proven to be very stable for us in many projects without generating any additional effort. Factorial&#39;s Frontend Lead Michael Großklaus has described more detailed technical details in a blog post: Mobile First versus Shared First CSS

Shared First

Shared-First CSS Approach

Make things, show them to people, learn from their feedback: Prototyping is a widely used method for quickly exploring technical and visual approaches to digital products and services. The method clearly emphasizes doing. It is common practice to create prototypes to explore options and possibilities before and during the development of a project. Today, software systems are used to provide a variety of functions and capabilities. Systems that do a lot for us are usually very complex. Due to their size or many unknown factors, extensive planning before development begins is hardly possible or simply not desired. By putting software prototypes into the hands of stakeholders and users at a very early stage, it is possible to collect valuable feedback. This helps to make better, more informed decisions and get risks under control.

Software Prototyping

Software reliability engineering is a set of principles and practices that uses aspects of software engineering and applies it to operating IT infrastructure.

Software Reliability Engineering (SRE)

Storybook is an open-source tool for building UI components and pages in isolation. It provides a development environment that runs outside of your main application, allowing developers to work on components independently and in isolation from business logic and data. The tool supports a wide range of frontend frameworks including Vue.js, React, Svelte, Angular, and more. It enables developers to create &quot;stories&quot; that showcase components in different states and configurations, making it easier to develop, test, and document UI components. Storybook is particularly valuable for building and maintaining design systems. It allows teams to create a living style guide where components are documented with examples, props, and usage guidelines. This documentation stays in sync with the actual code, reducing maintenance overhead. The tool includes addons for accessibility testing, visual regression testing, responsive design testing, and more. It integrates with popular testing tools and can be used to catch visual regressions and ensure components work correctly across different states and configurations. Storybook&#39;s ecosystem includes numerous addons that extend its functionality, from design tokens integration to automated visual testing. It can be deployed as a static site, making it easy to share component libraries with designers, product managers, and other stakeholders. With its focus on component-driven development and design system documentation, Storybook has become the industry standard for component development, especially in teams building reusable component libraries and design systems.

Storybook

Storybook Component Development Tool

Svelte is a JavaScript framework for building user interfaces (UIs). It&#39;s often categorized as a &quot;compile-time&quot; framework, which means that instead of shipping a bulky runtime library to the browser like other frameworks (e.g., React , Vue.js ), Svelte shifts most of its work to build-time. This approach results in smaller bundle sizes and faster performance. With Svelte, you write your components using a syntax that resembles standard HTML, CSS, and JavaScript. Then, during the build process, Svelte compiles these components into highly efficient JavaScript code that directly manipulates the Document Object Model (DOM) to update the UI when data changes. One of the key features of Svelte is its reactivity system. When the state of a component changes, Svelte automatically figures out what needs to be updated in the DOM, minimizing unnecessary re-renders and resulting in very efficient UI updates. Overall, Svelte offers a refreshing take on front-end development, emphasizing simplicity, performance, and a low learning curve. It&#39;s gained popularity among developers for its elegant approach and the advantages it brings in terms of performance and developer experience.

Svelte

Svelte Framework

Tailwind CSS is a utility-first CSS framework that provides a comprehensive set of low-level utility classes to build custom designs directly in HTML. Unlike traditional CSS frameworks that provide pre-built components, Tailwind gives developers the building blocks to create unique designs quickly. The framework promotes rapid development by eliminating the need to write custom CSS for common patterns. It uses a mobile-first responsive design approach and includes utilities for spacing, typography, colors, flexbox, grid, and more. Tailwind&#39;s JIT (Just-In-Time) compiler ensures that only the CSS classes actually used in the project are included in the final bundle, keeping file sizes small. Tailwind CSS has become the industry standard for modern web development, especially in the Laravel ecosystem where it&#39;s the default choice for new projects. It integrates seamlessly with popular frameworks like Vue.js, React, and Svelte, and works well with component-based architectures. The framework promotes consistency across projects through its design system approach. It&#39;s highly customizable through configuration files, allowing teams to define their own design tokens, colors, spacing scales, and breakpoints that match their brand guidelines. Tailwind&#39;s utility-first approach complements component-based development, making it easier to build and maintain design systems. It reduces context switching between HTML and CSS files, leading to faster development cycles and improved developer experience.

Technologies Library