Why SurrealDB is the one-stop shop database platform for AI-powered web development platform, SiteForge.
Operational overload is a common challenge many scale-ups and organisations face, particularly those with rapidly evolving products. It hinders innovation, slows development cycles, and increases maintenance costs. SiteForge, an emerging AI Content Copilot, is one company that experienced an operational overload.
We’ll explore how SiteForge successfully reduced its development cycle by 40%, and Queries by 20% and scaled back the number of backend APIs by 75%.
About SiteForge
SiteForge is an AI wireframe generator designed to help users quickly and efficiently create, modify, and iterate on web page wireframes and website plans. The platform leverages advanced AI to automate the wire-framing process, making it accessible to users with varying levels of technical expertise.
Albert Marashi, CEO and founder of SiteForge, has assembled a small yet ambitious team. Together, they help over 300 customers automate their content processes, tailor content to their brand identity, and create sitemaps that help them plan their websites in seconds.
Navigating complexities
SiteForge encountered various issues with their previous database solution, including the limited and complicated capabilities of the SDKs provided, such as incomplete JSONB querying and support.
The team had to run multiple queries in Supabase, whereas SurrealDB allowed them to do the same in one query on the client side, significantly improving developer productivity.
SiteForge needed to iterate quickly, update the schemas, and release product updates without worrying too much about migrations. This is why SiteForge selected SurrealDB.
Why SurrealDB?
Albert first became aware of SurrealDB after discovering a video by Fireship. This piqued their interest, especially in SurrealDB’s multi-model features, such as the ability to store JSON, retrieve nested JSON objects, graph relations, vector embeddings, etc.
As the SiteForge team continued to evaluate possible database alternatives, they considered SurrealDB’s features and how the database could potentially help them. They found considerable value in several key features of SurrealDB, specifically Record IDs, Record Relations, Nested Objects & Arrays, and Nested Queries, which have enabled them to turn complex information into clear insights, spot new opportunities and solve problems quicker.
“Nested Objects & Arrays is a feature that the SiteForge Engineering Team loves. The capability to store structured and nested JSON-like data is a must in SiteForge use cases. They require them to provide the rich functionality behind their AI wireframing block-based page builder.”
Improved development efficiency by 40%
The development cycle for database-related features has been reduced by at least 40%. SurrealDB’s flexible schema and intuitive query language have made it easy for the SiteForge team to add and support new features.
Minimising backend work by 75%
The combination of client-side database querying and authorisation means SiteForge can build far fewer backend APIs to handle feature functionality, effectively reducing backend APIs by 75%. This new paradigm of database-as-a-backend has been truly amazing for the team’s productivity and iteration speed.
Queries reduced by 20% with SurrealDB
The SiteForge team migrated away from Supabase and reduced the number of queries in their project by about 20% since they could merge operations due to the expressiveness of SurrealQL and its subqueries. They did all this in less than 24 hours, migrating all their customer data from Supabase to SurrealDB.
Optimised database architecture
They opted for SurrealDB default IDs rather than UUIDs because this improves database performance, reduces storage costs, simplifies data management, and enhances overall system efficiency. SiteForge now has faster application response times and a better user experience.
Future plans
The SiteForge team will use SurrealDB’s vector features to build more AI recommendation and memory systems in the future. They also plan to create website analytics systems for their future CMS using SurrealDB’s time series features and pre-computed table views. This will eliminate the need for a separate analytics system, reducing complexity and costs while providing real-time, scalable insights for their customers.
Transform your database management with SurrealDB.
The Featured Blog Posts series highlights posts from partners and members of the All Things Open community leading up to ATO 2024.