AI only works with architecture: watch the masterclass now 
Digital innovation in the building materials trade: AI image search for STARK Germany 
The right system for every application: Medienwerft Demo Days – See for yourself 
End2End solutions from UX to hosting: Benefit from the expertise of the FIS Group! 
Do you fit into our team? Discover all vacancies now and apply now! 
Ready for the shipyard? Find out what you can look forward to as an apprentice at Medienwerft! 
AI in e-commerce: discover possible use cases now! 
TCO halved: Successful replatforming for HABA Pro 
GENERATIVE AI – automated content
through LLMs & embedding models
Generative AI is revolutionizing the way content is created, processed and used. It is based on powerful models such as Large Language Models (LLMs), which are trained by very large amounts of data and can generate human-like texts. Embedding models are used to convert the texts into numerical vectors so that the AI can semantically understand and process content. Retrieval Augmented Generation (RAG) in turn combines AI-generated texts with external, up-to-date information from databases or the internet, which enables automated content generation as well as support from sales assistants in Assistant & Guided Selling.
Use data protection-compliant AI for content creation and sales support – our experts will be happy to support you!
Language models – in particular large language models (LLMs) – enable machines to understand and process language and generate texts themselves. In combination with generative AI, this opens up new possibilities in areas such as document analysis, dialog systems, content creation, semantic search and knowledge management.
For companies in e-commerce and wholesale, this means
More automation, better customer service, intelligent data processing and real innovation.
Why we’re the right partner for generative AI
- In-house research department with a focus on LLM applications
- Already implemented AI projects with generative AI in the wholesale and e-commerce environment
- In-house developed tools such as FISMWASP, Category Mapper or PDF Analyzer
- Expertise in data protection & governance, especially for sensitive data
- Technology-agnostic & customizable: open source models or API-based solutions – depending on your needs
What challenges can the technology solve for you?
Generative AI offers a wide range of options for efficiently overcoming complex challenges. It can generate creative content, automate processes and support problem-solving, for example through texts, images or code.
Whether the aim is to save time or to develop innovative approaches to various challenges, language models and generative AI can be used to solve a wide range of specific business problems:
Example: Companies receive large volumes of PDFs, emails or support tickets every day. Manual evaluation is time-consuming and error-prone.
Solution: Language models automatically analyze unstructured content – such as product data sheets, contract documents or support requests – and extract relevant information in a targeted manner.
Example: Employees cannot find important information quickly enough – for example on products, processes or contract conditions.
Solution: With RAG chatbots, we access internal sources (e.g. manuals, intranet) and provide precise answers in natural language.
Example: Product descriptions, category texts or email templates are created manually – this takes time and ties up resources.
Solution: Our language models automatically create SEO-optimized product descriptions, category texts and email templates that take the pressure off marketing and editorial teams.
Example: Users cannot find products or receive unsuitable suggestions in the online store.
Solution: Semantic search and intelligent recommendations based on embedding models improve search results and provide personalized suggestions that optimize the user experience.
What makes our AI service stand out?
We do not offer off-the-shelf AI, but tailor-made language model solutions – developed, integrated and supported by experienced AI teams. Pre-trained models are adapted to specific domains, e.g. technical language in wholesale, internal processes or product data. With this fine-tuning, your individual requirements can be optimally covered.
We are also a leader when it comes to data protection: thanks to hosting on our own servers in Germany, all of our AI solutions meet the strict requirements of the GDPR.
LLMs (Large Language Models)
Use of powerful models such as GPT, LLaMA, Mistral etc., on-premise or cloud-based depending on the use case
RAG (Retrieval-Augmented Generation)
Building contextual chatbots or assistance systems that access internal knowledge
Embeddings & vector search/FISMWASP
Proprietary solution “FISMWASP” for vectorization and semantic search in technical ticket data
Prompt Engineering & Tooling
Development of robust prompt strategies, evaluation and integration into existing tools and processes
Integration into existing platforms
e.g. PIM, ERP (SAP), e-commerce systems, intranet, CRM or customer portals
Chatbot for internal knowledge:
Previously: Support and sales teams had to manually access numerous documents and knowledge databases, which was often time-consuming and error-prone.
With AI: A RAG-based assistant accesses thousands of documents and internal data and immediately provides context-relevant answers – faster, more precisely and around the clock. This increases efficiency and improves customer service.
Automatic evaluation of PDFs:
Previously: Technical specifications in PDF form had to be checked manually and converted into structured data – a time-consuming and error-prone process.
With AI: LLM & PDF Analyzer work together to read technical documents, extract relevant information and convert it into structured master data. This saves companies time and improves data quality.
Product text generator:
Previously: The creation of product descriptions and category texts was a time-consuming process that often required manual revision and SEO optimization.
With AI: LLM models automatically generate SEO-optimized texts based on PIM data and target group profiles, which drastically reduces the effort and simultaneously increases visibility in search engines.
Technical ticket analyzer tool:
Previously: Support tickets were worked through manually, which slowed down processing and often meant that solutions were overlooked or not suggested quickly enough.
With AI: FISMWASP recognizes semantic similarities between support tickets and automatically suggests suitable solutions or FAQ entries – this speeds up the support process and ensures faster problem resolution.
Innovative and intelligent –
AI IMAGE SEARCH FOR STARK GERMANY
Intelligent product recognition with AI: STARK Germany relies on an AI-supported image search that enables customers to select products quickly and precisely. The technology opens up completely new ways to make the selection process in the building materials trade more intuitive and efficient.
How we work – our process model
Which processes can be improved with language models?
Selection of suitable LLMs, embedding solutions & orchestration
Rapid implementation of an MVP – directly on real data
Connection to existing systems, development of vector databases & APIs
Enablement of your team & continuous optimization
Blog posts about artificial intelligence
Automate your content creation – with our support and AI!
Get in touch with us: As an innovative AI service provider, Medienwerft offers you individually developed solutions, hosted in Germany in compliance with data protection regulations and perfectly tailored to your requirements. Make your company fit for future challenges!
Sönke Baumann
Head of New Business
040 31 77 99 0
sbaumann@medienwerft.de
✓ 15-30 minutes initial interview
✓ Discover Use Case Showroom
✓ Book AI-Demo
