AIMAX®
- Getting started is this easy
with the enterprise AI platform
From scratch to AI in days rather than months
The basic framework
- Bringing AI to staff
and enhancing organisational knowledge
The first step is to make the capabilities of AIMAX® available to all your staff.
The first benefit arises from the fact that AIMAX® makes your company’s knowledge more accessible – whilst, of course, taking data protection and data security into account.
Here, we’ll show you just how quickly and with how little effort you can start reaping the benefits of the AIMAX® Enterprise AI platform.
We’ll then show you how to get more out of AIMAX®, right through to how you can fully exploit the potential of artificial intelligence.
It’s that easy to get started!
Install AIMAX®
Integrate the local AIMAX® OnPremise version into your network, or let us set up AIMAX® for you in a private cloud.
Both options have their merits and can be scaled as required.
The OnPremise version of the AIMAX® Enterprise AI platform is, for example, compatible with Apple® Mac Studio® as well as NVIDIA DGX Spark™ and equivalent devices from other manufacturers. In doing so, AIMAX® harnesses the performance and efficiency of these devices to your advantage, delivering outstanding value for money whilst ensuring excellent sustainability.
We’d be happy to discuss the necessary hardware requirements with you and/or demonstrate the capabilities of AIMAX®.
Create a user
Add the required users to your AIMAX® stack. This can be done manually via the AIMAX® backend by entering a username and a valid licence key, or by synchronising with your Active Directory.
Detailed settings can be configured for each user, such as information about their personality, the desired AI behaviour, their preferred AI model and the assigned user roles. This enables targeted and transparent management of user profiles.
When using the Microsoft Teams plugin, users are created automatically the first time AIMAX® is used.
Assigning roles and permissions
To ensure that each user can only access the data, interfaces and agents intended for them, create the appropriate access and authorisation structures in Role and Authorisation Management.
A role represents a specific permissions group in this context. You can create any roles or permissions groups as required.
Tip: A user, etc., can of course be assigned multiple roles.
In larger organisations, roles can be assigned by synchronising with Active Directory and mapping the permission groups.
Integrating internal knowledge
In the AIMAX® stack, Knowledge Vaults are used to connect different information sources.
There are different types of Knowledge Vaults for different data sources, each specifically tailored to the respective data source.
File-based Knowledge Vault types (Document Storage and WebDAV) support a wide range of file formats, including:
- PDF files
- Office formats: Word, PowerPoint, Excel
- Images (including EXIF data and descriptions)
- Web and data formats: HTML, JSON, XML, CSV
- EPub format
- ZIP archives
- …
In principle, all documents and information remain in their original locations; only a cache is created. This enables the AI models to access this knowledge quickly and reliably.
Configuring channels
Before you can get started with your AIMAX® in the form of a Company-GPT, you must first configure the channels through which you wish to communicate with or work using the AI.
At this stage, you have the following options to choose from, which you can enable or, in some cases, configure in more detail:
- WebChat
- Microsoft Teams
- Microsoft Outlook 365
- Zendesk
- OpenAI-compliant API
Using AIMAX® in WebChat or Microsoft Teams
You can start using AIMAX® straight away. To do so, you can use WebChat or the Microsoft Teams app, which you can install in just a few clicks. All you need is the licence key, which can be found in the user management section.
Furthermore, you can already start using the service via the OpenAI-compliant AIMAX® API. You’ll also find the relevant authentication token (Bearer) in your user settings.
The AI infrastructure solution
for greater productivity!
Contact us now.
AIMAX Business Solutions combines outstanding software made in Germany with first-class service. With unique AI systems designed with the user in mind, we unlock new potential within your business.
The key feature: 100% data security and GDPR compliance thanks to local AI.
Create AI agents and integrate other systems
- get more out of AIMAX®
AIMAX® can do much more than simply answer your questions and serve as a knowledge base for your staff.
By integrating external systems, the agent platform can also provide information based on real-time data or trigger processes based on data prepared in advance. AI agents defined by you can thus take on specialised tasks and systematically achieve the objectives set for them.
Free yourself from tedious routine tasks and create space for value-adding activities. Let’s get started!
Connect further systems
Your AIMAX® stack comes with a wide range of built-in connectors. These are used to connect to external functionalities and can be utilised to retrieve data or information in real time from other systems, or to trigger external functionalities.
Your AIMAX® stack offers the following options:
- Connection to any MCP server (via SSE or Streamable HTTP)
- Invoking webhooks from third-party systems
- Connection to any REST APIs – no programming knowledge required (no-code)
For all these integrations, you can specify whether a prior manual check and possible correction of the data should be ensured by the integrated ‘human-in-the-loop’ functionality before the call is made.
Create AI agents
Your AIMAX® stack offers the option to define your own AI agents. These are assistants which can independently plan how to achieve the objectives set for them, utilise tools and thus solve complex tasks.
Managing the AI agents allows you, amongst other things, to set goals, define clear boundaries or constraints, and specify the behaviour or personality of the AI agent.
Using ‘Classification’ and corresponding ‘Classification Examples’, you can specify when, or in which use cases, your AI agent should be deployed.
You can also assign a selection of tools to your AI agent, which may be used to achieve the objectives.
Setting up a cloud AI provider
If you wish, you can also add cloud AI providers and their AI models to your AIMAX® stack for use, in addition to the local AI models. To do this, use the API credentials provided by the respective provider. The following cloud AI providers are supported by AIMAX® out of the box:
- Azure Foundation
- OpenAI
- Anthropic
- Mistral
If required, we can also set you up to connect to other cloud AI providers.
“The AIMAX® Enterprise AI platform was developed from the outset to offer the greatest possible ease of use and the fastest return on investment. This naturally includes seamless integration into your existing system landscape and processes.
Please feel free to contact me if you have any questions about compatibility with your existing legacy applications.”
With ontologies and tool builders
make the most of AIMAX®’s capabilities
Now that the basic setup is complete, any additional systems have been integrated where necessary, and useful AI agents have been created, it is time to really make the most of your enterprise AI platform’s capabilities.
Using ontologies, you can significantly improve the quality of your AI’s responses – particularly within your specialist field – whilst the Tool Builder helps you automate complex tasks where AI and other systems need to work hand in hand.
For example, create a tool for automatically generating suggested email replies to your customers’ enquiries. Using the Tool Builder, you can set up a workflow that analyses the enquiry when an email is received and generates a suggested reply. The right ontology ensures that the generated replies are of a high standard. And a ‘Second Reviser’ integrated into the workflow can perform an AI-based second check to verify for errors before an employee receives the result.
So what are you waiting for?
Create tools or automations
The Tool Builder integrated into AIMAX® features a user-friendly interface. It is essentially designed as a no-code application, but also supports the integration of code snippets where required.
With around 500 ready-made ‘building blocks’ (Pieces), the Tool Builder offers a comprehensive set of functions, such as accessing spreadsheets or creating to-do lists. It also enables integration with numerous cloud and online applications and platforms, including Google services and social media platforms such as Instagram and LinkedIn.
As well as being triggered by AI agents, the tools can also be launched on an event-driven basis (time, incoming emails, requests, data changes, etc.), thereby offering a wide range of potential applications for automation.
Integrate Codex
To make even more direct use of artificial intelligence’s capabilities in your working environment, AIMAX® can also be connected to and enhanced by Codex.
This enables direct integration into your workplace environment, allowing you, for example, to create or edit documents (PDF documents, Microsoft PowerPoint files, etc.). You will thus receive support even with complex tasks.
Creating ontologies
In addition to the method of knowledge enrichment based purely on text search, the AIMAX® stack also includes the option to create knowledge graphs or a GraphRAG.
In this context, AIMAX® enables the identification of connections within the data that are not immediately obvious. It can be likened, to some extent, to reading ‘between the lines’.
For example, the system recognises the connection between the name ‘Henryk Liebezeit’, the associated email address ‘h.liebezeit@get-aimax.de’ and an internal abbreviation formed from the initial letters ‘HLI’. If a search is then carried out for a topic in connection with the name, information is also found where the topic appears in conjunction with the abbreviation.
This approach further significantly improves the quality of the answers provided by the AI.
Manage AI models
In addition to the features already outlined, AIMAX® offers you significantly more extensive options for AI model management.
As well as the option to globally deactivate specific AI models, you can also deactivate models on a user- or role-specific basis, thereby restricting access to certain models to specific user groups, for example.
To achieve the best possible results, you can also configure different system prompts for the various models, depending on the channel and the model. This allows you to compensate for the specific behaviour of certain models or, for example, to enable specific output formats (HTML, Markdown, etc.) on a channel-by-channel basis.
Manage API gateways
To enable differentiated API access, you can create OpenAI-compliant API accesses within the user context in AIMAX®.
Within this framework, you can restrict both the available AI models and the tools authorised for use for each API access. This serves primarily to minimise potential security vulnerabilities and to grant each use case only the rights it actually requires.
In addition, access to these API entries can also be time-limited to prevent them from remaining open for unnecessarily long periods.
Other relevant topics
AIMAX® platform architecture • AIMAX® core features • AIMAX® RAG types • AIMAX® Quick Start • An overview of our AI agents
Some of our valued reference clients
