Skip to content

 

Build Insights,
Not Infrastructure

We’ve taken care of all the essential building blocks so you don’t need to build them yourself. Our SmartSpace infrastructure includes data connectors, data containers, vectorisation, access control, developer tools, multi-agent collaboration, and an LLM marketplace. 

Our platform provides a secure, scalable foundation for deploying generative AI solutions self-hosted in your own environment.

build-insights
Large Language Models we support
  • Large Language Models we support
  • Gemini
  • Hugging Face
  • Open Ai
  • Anthropic

SmartSpace Infrastructure

git-fork

Data Sources and Connectors

Connectors in SmartSpace allow you to sync and organise data from multiple sources seamlessly. Whether it’s from Azure Storage, SharePoint, or HubSpot, our native connectors hold the necessary credentials for data access and keep your data flowing smoothly into your workspaces. 

Research

Data Spaces

Data Spaces are containers for combining, storing, and processing data from multiple platforms for secure use within a Workspace. Select the data you need from each connector, and our platform sets up a continuous sync job. As new data comes in, it’s can be enriched using LLMs and automatically vectorized and stored in our vector database.

Group (4)

Large Language Models

Our platform offers connectors to all leading public (or frontier) models, ensuring seamless integration with top-performing LLMs. Additionally, we can host open-source models, such as those found on Hugging Face, providing flexibility and choice for your AI projects. This enables you to leverage the best models available, tailored to your specific needs.


Programing, Data

Data Ingestion

SmartSpace.ai features a data ingestion engine that can be configured to sync data from external sources into the SmartSpace platform. A delta copy of the data is taken every few minutes, and the data is vectorized and stored for use by workspaces.

server-database-data-style-2-send-fast

Data Enrichment

Data retrieval for use in workspaces is greatly enhanced by enriching ingested data through various methods, such as tagging, entity recognition, named entity extraction, and key date extraction, among others.

Group (14)

Data Vectorisation

Data is vectorized and stored for easy retrieval, optimizing both performance and accessibility. Storage options, such as chunk sizes and vectorization strategies, can be configured to suit specific needs and use cases. This flexibility ensures that data handling is efficient and tailored to individual workspace requirements.

Group 54

Instead of spending four weeks building infrastructure, spend that time bringing multiple use cases to life.

Group (17)

Access Controls

Access controls are integrated into workspaces and utilize your Entra (SSO) configuration. You can add individuals, groups, or principal accounts for API integration, ensuring secure and tailored access management.

Group (18)

Developer Tools and APIs

For advanced use cases, developer tools and SDKs are available to create custom Python blocks for use in workflows. You can debug these from the Admin section or in a local IDE. An extensive list of APIs is available to suit your specific needs and build the required integrations.

Group (19)

Deployment on MS Azure

Hosted in your own secure Azure environment, rest assured that nothing leaves your infrastructure without your permission. SmartSpace can be easily installed from the Azure Marketplace, providing a seamless and secure deployment experience.

Deploy today!

SmartSpace is your ideal partner for building and implementing AI quickly and effectively. Our plug- and-play infrastructure transitions you from concept to execution in minutes. This allows you to get to value quicker, safer, and with seamless iteration at your fingertips.