- SurrealDB is an AI-native, multi-model database that provides Agentic Memory – the persistent context and reasoning layer – for modern apps and autonomous agents.
- It unites document, graph, relational and vector data with built-in inference, running seamlessly from cloud to edge.
- It halves infrastructure complexity and cost for AI products.
Website: https://surrealdb.com/
What Is SurrealDB?
Tobie Morgan Hitchcock, CEO of SurrealDB, reflects that in the first months of 2023, “we were providing a cutting-edge multi-model database, centred around schema flexibility and real-time graph links. With the arrival of AI, everything has moved up a gear: AI needs memory. SurrealDB now gives agents an Agentic Memory – a single place to store documents, graph edges and vectors, run on-device inference, then sync to the cloud so they can reason even offline.”
The journey from the very early days of SurrealDB (2018) has been a challenge, following Tobie’s experience of working on varied SaaS products, which is when he started to envisage what a brand new database concept could look like. “We were working on a complicated graph/analytics/visualisation platform for golf-course analytics, using multiple databases (Influxdb, OrientDB, DynamoDB, MySQL) and a complicated API layer sitting inside CoreOS. We thought that there had to be a better system, but there wasn’t. So, we decided to make one, and SurrealDB was conceived.
Amongst other things, we separated the storage from the compute, which enabled us to run SurrealDB as an embedded database, but also as a cloud database where you can query and distribute your data in many different ways and scale as you need to.” The aim was always to stress simplicity, and consequent time saving (and cost reduction).
The interest shown by developers was immediate and dramatic: SurrealDB became the fastest-growing database in history, amassing 10,000 stars on Github in 4 weeks; it has now reached just under 30,000, with over 2.10m downloads. Following huge support from the developer community from its earliest days, SurrealDB is now demonstrating similar appeal for a very wide range of industries: finance and fintech, defence and aerospace, healthcare, retail and e-commerce, energy and manufacturing, and gaming and entertainment. Each uses it in different ways, but cost reduction and ROI are still imperative today.
Want to be featured as TechRound’s Startup of the Week? Find out more about this weekly feature, and how to get involved, here.
Is There A Market for the Retail and E-Commerce Sector?
One particular sector which is experiencing the benefits SurrealDB can offer is customer support in retail and e-commerce, as companies seek dynamic pricing, live inventory, and Al-driven recommendations in milliseconds – in particular, real-time personalisation at scale. The USA luxury department store chain Saks Fifth Avenue has 5 million active customers and processes 45 million recommendation queries monthly.
Customer preferences, purchase history, browsing patterns and product relationships had been spread across siloed systems, making unified analysis impossible. Consolidation of customer data from multiple touchpoints into a single, scalable platform, eliminated data silos and enabled comprehensive customer insights. The integration of large language models (LLMs) with SurrealDB’s vector capabilities were able to deliver contextually aware, personalised product recommendations that adapted to customer behaviour in real-time.
Saks saw a 167% increase in conversion rates, and a boost of 30% in customer retention. Samsung Ads had similar issues – with customer profiles and behaviour data stored in siloed systems, and an inability to segment audiences effectively resulting in a high ad spend with suboptimal returns. Samsung Ads were able to achieve a 25% ROI in their precise audience targeting, saving millions in annual ad spend; and also cut operational costs by 30% by eliminating the need for multiple analytics platforms.
How To Access SurrealDB
SurrealDB is available from surrealdb.com. The free level is available for prototypes, side projects and getting started, and includes community support, an integrated AI copilot, and comprehensive documentation, guides and tutorials.
