Consulting and support

Alcilla can begin locally on a single workstation or be developed into a shared enterprise solution. You do not need your own AI department to use it.

We support both companies with existing infrastructure and organizations using local AI models for the first time.

Consulting before introduction

Requirements workshop

Together, we clarify:

• which workflows should be improved,
• which data may be processed,
• which functions are actually required,
• which systems need to be connected,
• whether local or centralized operation is appropriate,
• which success criteria should apply to a pilot project.

Architecture and infrastructure

We assess:

• existing workstations, servers, or virtual machines,
• CPU, memory, and GPU resources,
• network, proxy, and certificates,
• the identity provider,
• storage requirements,
• backup and operational requirements.

The result is an understandable target architecture with a clear division of responsibilities.

Installation and setup

Possible services include:

• installing Alcilla,
• setting up Ollama,
• loading and testing suitable models,
• deploying optional services such as Docling, SearXNG, or transcription,
• configuring email and calendars,
• setting up the knowledge index,
• connecting a central sign-in system,
• configuring centralized operation.

Model and hardware consulting

A larger model is not automatically a better model. We compare quality, speed, and resource requirements using your specific use cases.

We can:

• assess existing hardware,
• select suitable models,
• compare local and central variants,
• recommend GPU or server sizing,
• use typical questions and documents as a test dataset.

Training and enablement

We offer training for:

• users,
• administrators,
• IT operations,
• data protection and information security teams,
• subject-matter champions.

Operation and support

Depending on the agreement, we can assist with:

• updates,
• model changes and reindexing,
• troubleshooting,
• adapting integrations,
• performance problems,
• increasing the number of users,
• documentation and handover to operations.