WZ-IT develops, reviews, migrates and operates AI applications. We connect custom software, RAG and automation with local models, controlled infrastructure and a reliable deployment and operations path.
Make company knowledge usable with sources and permissions.
Integrate AI features, agents and automation into existing systems.
GPU infrastructure, monitoring and operations from one team.
Companies worldwide trust WZ-IT
Assessment, validation, production applications, platforms and operations each have their own scope. Start where your project stands today.
Assess one business process and decide whether AI, conventional automation or no implementation is the appropriate next step.
€3,900 excl. VAT
View scopePlan business opportunities, target architecture and the operating model for a wider AI initiative.
from €8,900 excl. VAT
View scopeEvaluate a scoped knowledge collection against agreed reference questions and verifiable sources.
from €9,900 excl. VAT
View scopeIntroduce an internal knowledge assistant into production on a suitable existing platform.
from €34,900 excl. VAT
View scopeIntroduce one workflow with defined tools, verified permissions and controlled actions into production.
from €39,900 excl. VAT
View scopeBuild the shared technical foundation for controllable AI applications in your target environment.
from €79,000 excl. VAT
View scopeRegularly assess answer quality, errors, latency and cost for one AI application.
from €1,990 excl. VAT / month
View scopeSix clear entry points, one principle: we choose the smallest viable scope and connect implementation with operations only where the process requires it.
| Your situation | The system that fits | When it does not fit |
|---|---|---|
| A recurring process consumes time, but it is unclear whether AI, fixed rules or conventional automation are appropriate. | AI process assessmentOne process, its data, systems, risks and a viable pilot path. | If the workflow, data and target cannot yet be described concretely, the process itself must be clarified first. |
| Visitors land on the site and cannot tell which of our offerings matches what they want to do. | Offer finderChat widget on your website. | If your offering is three clear services, good navigation is enough. |
| Staff search historic quotations, copy recurring line items and then enter them again in the accounting system. | AI quotation assistantInternal working system for reviewable quotation drafts. | If the business only sells a few standardised products, a maintained catalogue in the accounting system is often sufficient. |
| We hold a large corpus. Readers need the authoritative passage, not a summary. | Knowledge navigatorGuided search with source and version. | As long as valid and superseded versions cannot be told apart in the corpus, tidying up comes first - not a navigator on top of it. |
| Staff search internal documents. Who may see what has to hold technically, not by instruction. | Internal assistantRAG on your documents, permissions before retrieval. | If it comes down to a few recurring questions, an FAQ or a form is cheaper. |
| A task should not only be answered but carried out - open a ticket, book an appointment, query data. | AI agentsActions in your systems: tickets, workflows, integrations. | As long as nobody can state what the agent is allowed to have broken when it goes wrong, it is too early. |
The AI Cube provides a prepared platform for a local start. When standard hardware or on-site operation is not the right fit, the corresponding expansion paths are: AI Cube, GPU servers or LLM hosting.
Four patterns we see again and again - and that one more tool rarely solves.
Prototypes impress, but never make it into secure day-to-day operation.
Documents, wikis, tickets, databases and APIs are not usable as AI context.
Chatbots sit next to the processes instead of being embedded in portals, workflows and systems.
Roles, sources, logging, monitoring, cost control and updates are planned too late.
Production AI needs four layers designed together: application, knowledge, inference and operations. We build and operate them as a whole.
Assistants, agents, UX, workflows, portals and business apps.
RAG, data sources, embeddings, Qdrant, sources and permissions.
Ollama, vLLM, models, GPU sizing and LiteLLM as the gateway.
Langfuse, monitoring, evaluation, cost control, updates, security and SLA.
The AI Cube brings local AI into your organisation as a prepared platform in under two weeks. We build internal assistants, knowledge systems and custom integrations on top without reinventing the infrastructure for every use case.
Delivered ready to use within 10 working days
€6,490 excl. VAT one-time
Device, setup, briefing and shipping
€349.90 excl. VAT / month
AI Cube Care · first year accompanied, then cancellable monthly
RAG, data sources, integrations, secure connectivity and custom workflows are scoped separately.
AI Cube Care keeps the device running: a monthly maintenance window, tested model updates with a rollback path, round-the-clock monitoring, ticket support and warranty handling. For confidentiality professionals we add a data processing agreement under Art. 28 GDPR and the written secrecy obligation under Section 203 (4) of the German Criminal Code, including instruction on the criminal consequences; on request remote access stays switched off by default.
Make knowledge accessible, support sales and purchasing or process recurring work: each offer connects a specific workflow with suitable data, approvals and operations.
Make expertise, company documents and service cases accessible in daily work.
Prepare quotation drafts and evidence-backed responses using existing company data.
Check supplier confirmations and prepare product data with controlled approvals.
Capture and validate incoming orders, documents and messages and hand them to the relevant systems.
Deploy models locally, in a hybrid setup or suitable cloud and support the application.
Assess further use cases and connect them to APIs, approvals and existing software.
The services can be commissioned individually or as consecutive steps. Target environment, operating location and responsibility model are then defined around the application.
Develop
Develop AI features, RAG, agents, APIs and user interfaces from scratch or continue the product together with your team.
View AI software developmentAssess
Assess AI-generated applications for code, authentication, data, security, tests, deployment and operations before go-live.
View the auditReview continuously
After a confirmed baseline, have changes, CVEs and release risks reviewed and prioritised by a human on an ongoing basis.
View the review retainerMake production-ready
Take over MVPs from Lovable, Bolt, v0, Replit or Base44, resolve prioritised findings and establish a controlled production path.
View the takeover processMigrate
Move Supabase Cloud, Firebase, Lovable or existing backends to a suitable Supabase target environment together with application, auth, storage and functions.
View the migration serviceAfter go-live
Depending on the system, WZ-IT assumes the agreed application, platform or infrastructure operations. Your team can continue development in a co-managed model while deployments, monitoring, backups, updates and response are assigned clearly.
Concrete entry points at the intersection of business processes, AI, software and operations.
Answers with citations from your own documents and systems - with roles and permissions.
Connect and prepare distributed knowledge and guide visitors to the right answer - with source and version.
Classify, extract and validate emails, PDFs, scans and forms and hand off structured results.
Validate customer orders against master and mandatory data and prepare them for ERP or CRM.
Classify, prioritise and route requests and prepare response drafts.
Assess workflow, data, systems, risks and a realistic pilot path before implementation.
Take over AI-generated MVPs, harden them and make them operable.
Control traces, quality, retrieval, cost, latency and changes on an ongoing basis.
We pick components by use case, data protection, load and operating model - not by hype. Open source provides control, but the stack only becomes productive through integration, permissions, monitoring and operations.
Local inference engine
Run and manage models locally and expose them through APIs.
High-throughput inference
Tensor parallelism and OpenAI-compatible APIs for many concurrent users.
Chat and admin interface
User interface for local models, teams, tools and document chat.
Enterprise RAG app
RAG system for documents, workspaces, agents and knowledge bases.
Multi-LLM gateway
Central API layer for routing, budgets, keys and provider fallbacks.
LLM observability
Tracing, prompt management, evaluations, cost and quality control.
Vector database
Semantic search and retrieval layer for production RAG systems.
Automation building block
Connect processes, APIs and LLMs into controlled workflows.
We do not plan AI operations in theory. WZ-IT runs cloud, open-source and software stacks in production, plus IoT and remote-site systems for real sites such as ABCO Water Systems and nextGYM. That operations experience feeds into local and hybrid AI systems.
Not every use case needs dedicated GPUs. The relevant factors are data protection, latency, cost, model size and operations responsibility. We size the stack to the risk and workload.
Enquiry
Describe the objective, data, and existing systems. We assess the solution, infrastructure, and next step.
Runs where you want
Answers to the most important questions
Recurring workflows with emails, documents, validation rules and clear target systems are often suitable. Examples include document processing, order entry, ticket and inbox automation, and custom workflows. The AI Opportunity Check first determines whether AI, fixed rules or conventional automation fit.
Assessment, validation, production applications, platforms and operations each have their own scope. Start where your project stands today.
Document processing extracts and validates data from different inputs. Automated order entry processes incoming customer orders for ERP or CRM. The quotation assistant instead uses your product and historic data to prepare new outgoing quotation drafts.
Retrieval-augmented generation connects the language model with your documents: before each answer, relevant passages are retrieved and provided. The system answers with your knowledge without training a model - and permissions can be enforced technically before the search, not just in the prompt.
Local or controlled European model operations can be suitable where data paths, model choice, latency or running costs have particular requirements. The right architecture depends on the use case, data categories, model size, concurrency and integrations.
Managed AI Server covers the underlying technology such as GPU, operating system, drivers, containers and inference runtime. Managed LLMOps covers the application above it: traces, quality, retrieval, cost, latency and changes to models, prompts or knowledge sources.
No. Local operation can constrain data paths and improve technical control. Purpose, legal basis, user rights, deletion, human oversight and the applicable risk framework still need to be assessed and documented for the concrete use case.
No risk: worst case, you leave with a clearer understanding of your project than before.
“WZ-IT's advice on our Azure migration was technically sound and completely non-binding right from the intro call - we took away a great deal.”
From local AI integration to architecture, data sovereignty and ongoing operations.
“WZ-IT moved our studio infrastructure from decentralised individual devices to a central platform: every site is securely connected via VPN, new devices are onboarded automatically and an entire site is provisioned from a template, without manual steps on location. What impressed me most is the breadth and depth of their knowledge: Timo and Robin are not a typical IT provider who sets up a server and leaves. The two of them think their way into highly complex infrastructure and software topics, work through every requirement we put in front of them, and build networking, provisioning and operations so that everything fits together in the end. WZ-IT is an excellent partner for complex software, network and architecture projects.”
Steve Kirchner
Managing Director, nextGYM GmbH
“While looking for a partner to integrate a local AI solution, we came across Timo and Robin - and could not have made a better decision. They quickly understood our requirements and solved technical challenges competently and pragmatically. We were particularly impressed by the straightforward collaboration, their high level of expertise and the speed of implementation. We are delighted to have Timo and Robin at our side as reliable business partners and look forward to our next projects together. (translated)”
Henrik Jeche
Head of IT Administration
ml&s - full-service provider for the electronics industry.
“Communication with WZ-IT was open, friendly and professional from the very beginning. Mr. Wevelsiep and Mr. Zins responded to every one of our questions patiently, promptly and respectfully, and contributed their own suggestions and good ideas on how to modernize our server infrastructure and adapt it even better to our needs. Throughout, WZ-IT consistently kept our need for data sovereignty and GDPR compliance in mind. Even short-notice adjustments and change requests from our side were implemented quickly and patiently and integrated consistently into the overall concept. From planning to the final handover, every step was documented transparently and comprehensively - so if a provider change should ever become necessary, the documentation would let us onboard a successor into our infrastructure very quickly. We are extremely satisfied with WZ-IT and hope to continue working together for a long time.”
Christoph Mußmann
Project Officer, DGHO e.V.
DGHO - German Society for Hematology and Medical Oncology.
“We have had very good experiences with Mr. Wevelsiep and WZ-IT. The consultation was professional, clearly understandable, and at fair prices. The team not only implemented our requirements but also thought along and proactively. Instead of just processing individual tasks, they provided us with well-founded explanations that strengthened our own understanding. WZ-IT took a lot of pressure off us with their structured approach - that was exactly what we needed and is the reason why we keep coming back. (translated)”
Matthias Zimmermann
Managing Director, Annota GmbH
Annota - healthtech startup with an AI documentation assistant.
“I got to know Timo and Robin as dedicated and professional partners who make a lot of things possible for their clients that initially seem impossible. I felt very well advised. The services offered have thoroughly convinced me. (translated)”
Andrea Pawlowski
Communications Manager, Golem.de
Golem.de - leading German IT news outlet.
“Professional, honest and technically thoroughly sound: WZ-IT set up our Proxmox and backup infrastructure securely and future-proof. The consulting from Timo and Robin was objective and needs-driven, the implementation smooth and the final documentation exemplary. You can tell immediately that business-critical infrastructure is thought through holistically from the outset here and implemented with clear ownership.”
Pascal Block
Project Manager, Stadtwerke Brühl GmbH
Stadtwerke Brühl - municipal energy and infrastructure utility.
“With Timo and Robin, you're not only on the safe side technically - you also get the best human support! Whether it's quick help in everyday life or complex IT solutions: the guys from WZ-IT think along with you, act quickly and speak a language you understand. The collaboration is uncomplicated, reliable and always on an equal footing. That makes IT fun - and above all: it works! Big thank you to the team! (translated)”
Sonja Aßer
Data Manager, ARGE
ARGE Neue Medien - master data & digital standards for plumbing and building technology.
“I recently worked with Timo and the WZ-IT team, and honestly, it turned out to be one of the best tech decisions I have made for my business. Right from the start, Timo took the time to walk me through every step in a simple and calm way. No matter how many questions I had, he never rushed me. The results speak for themselves. With WZ-IT, we reduced our monthly expenses from $1,300 down to $250. This was a huge win for us.”
Aleksandr Shuliko
CTO, EVA Real Estate
EVA Real Estate - leading real-estate agency in Dubai.
“WZ-IT manages our Proxmox cluster reliably and professionally. The team handles continuous monitoring and regular updates for us and responds very quickly to any issues or inquiries. They also configure new nodes, systems, and applications that we need to add to our cluster. With WZ-IT's proactive support, our cluster and the business-critical applications running on it remain stable, and high availability is consistently ensured. We value the professional collaboration and the noticeable relief it brings to our daily operations.”
Pascal Hakkers
Aphy AG
Aphy AG - AI platform for hotel back-office.
“WZ-IT provided very competent support in implementing our server and virtualization infrastructure. As part of the project, a Proxmox-based virtualization environment was set up along with a virtual machine for our application systems. Additionally, a structured backup concept was implemented. A particular focus was on security: the setup of a VPN tunnel and the clean implementation of encryption and access concepts were professionally executed. The structured consulting as well as the reliable and swift implementation deserve special mention. We are very satisfied with the services provided and are happy to recommend WZ-IT.”
John Hellmerichs
Managing Director, AInergy GmbH
AInergy - consultancy for energy and facility management.
“Timo and Robin from WZ-IT set up a RocketChat server for us - and I couldn't be more satisfied! From the initial consultation to the final implementation, everything was absolutely professional, efficient, and to my complete satisfaction. I particularly appreciate the clear communication, transparent pricing, and the comprehensive expertise that both bring to the table. Even after the setup, they take care of the maintenance, which frees up my time enormously and allows me to focus on other important areas of my business - with the good feeling that our IT is in the best hands. I can recommend WZ-IT without reservation and look forward to continuing our collaboration! (translated)”
Sebastian Maier
Managing Director, Yonju GmbH
Yonju - business coaching and AI products.
“Counting on WZ-IT team was crucial, their expertise and solutions gave us the pace to deploy in production our services, even suggesting and performing improvements over our configuration and setup. We expect to keep counting on them for continuous maintenance of our services and implementation of new solutions.”
Gabriel Sanz Señor
Managing Director, Odiseo Solutions
Odiseo Solutions - software engineering for the space industry.
“WZ-IT supported us professionally with the provisioning and setup of NetBird. Alongside the technical implementation, we also received well-founded advice on connecting our Kubernetes clusters. The recommendations were concrete and directly actionable. The collaboration was consistently reliable, fast and on equal terms. We can recommend WZ-IT without reservation.”
Kai Clauß
Head of Development, SweetConnect GmbH
SweetConnect - cross-manufacturer Industry 4.0 platform for confectionery production.
14.09.2026
Public authorities manage large collections that have grown over years: decrees and circulars, procedural instructions, technical reports, expert opinions, statements, minutes. People looking for a...
12.09.2026
Anyone planning a RAG knowledge base soon runs into the question of whether the system falls under the high-risk rules of the EU AI Act....
07.09.2026
Since 1 January 2025 every German company has to be able to receive e-invoices. Part of the incoming documents now arrive as structured XML and...
24.05.2026
Anyone bringing AI into production business processes quickly faces an uncomfortable question: what is actually happening in there? Which prompt went to which model, why...
10.05.2026
Three unauthenticated API calls. No login, no exploit framework, no privilege escalation. Three POST requests to a default port, and the machine's memory is on...
05.05.2026
The EU AI Act's high-risk obligations do not start on 2 August 2026. Stand-alone high-risk systems under Annex III now have until 2 December 2027,...
24.11.2025
OpenAI released GPT-OSS 120B as an open-weight reasoning model on 5 August 2025. Its native MXFP4 quantisation allows OpenAI to position the model for a...
Whether a specific IT challenge or just an idea - we look forward to the exchange. In a brief conversation, we'll evaluate together if and how your project fits with WZ-IT.
Companies worldwide trust WZ-IT
