alangu develops an AI-based platform for the automated translation of written and spoken language into German Sign Language (DGS) using digital avatars. The goal is to enable barrier-free communication for deaf and hard-of-hearing people wherever information is currently provided primarily in text- or speech-based form, both digitally and in physical environments.
The core of the business model is a scalable Software-as-a-Service solution. Companies and public institutions integrate alangu’s technology via web interfaces or APIs into existing systems such as websites, customer portals, service terminals, or internal applications. Usage is based on license models, complemented by project-based revenues for setup, customization, system integration, and quality assurance.
Technologically, alangu combines machine learning, linguistic knowledge bases, and multimodal corpus data, developed and validated in close collaboration with deaf experts. A strong focus is placed on quality, transparency, and ethical AI development, including human-in-the-loop approaches and continuous evaluation.
The business model addresses a rapidly growing market at the intersection of digitalization, artificial intelligence, and inclusion. Regulatory requirements for accessibility (e.g. BITV 2.0, BFSG, European Accessibility Act) further increase demand. At the same time, alangu creates social impact by improving digital participation for millions of deaf people and by creating qualified employment opportunities for deaf professionals.
alangu thus combines economic scalability with measurable social value and positions itself as a European technology leader for AI-based sign language solutions.

ByteIT is an AI Deep-Tech university spinoff addressing a critical bottleneck in AI customization: loading complex, unstructured documents into Large Language Models (LLMs) with high accuracy. Our solution improves LLM ingestion precision by up to 26% at reducing processing cost of under $0.01 per page. This is achieved through novel vision-model-based document ingestion platform that “gives LLMs glasses” – enabling them to see and interpret dense, visual, or unstructured documents as accurately as a human.

We operate in a fast-growing global market as enterprises race to deploy custom AI solutions. ByteIT aspires to be Europe’s first mover in this space, targeting a €20 billion opportunity in intelligent document processing for AI. U.S.-based peers (e.g. Unstructured, Reducto) demonstrated market traction by reaching seven-figure revenues within one year, validating the demand. ByteIT’s competitive edge lies in our superior precision and ultra-low cost: our benchmarks show order-of-magnitude cost savings and significantly higher extraction precision compared to existing tools. This translates into strong value for customers (better AI outcomes at lower cost) and high-margin scalability for the business.

* Why – The Problem We Refuse to Accept
European SMEs operate in increasingly volatile markets, yet many still rely on Excel spreadsheets and gut feeling to forecast demand. Inaccurate planning leads to excess inventory, stock-outs, and wasted capital. We refuse to accept that advanced forecasting should be reserved for large corporations.

* What – Accurate Forecasts Without Complexity
foresAIght delivers highly accurate, product-level demand forecasts on a daily, weekly, or monthly basis. Companies receive regularly updated forecasts that directly support production, procurement, and inventory planning, without requiring changes to existing systems or operational processes.

* How – The Technology That Makes It Possible
We combine proprietary meta-learning architecture with more than 1,500 external demand drivers, including weather data, macroeconomic indicators, and sector-specific trends. For each SKU, the system dynamically selects the best-performing model and continuously adapts to changing demand patterns.

* Who – The Companies We Build For
We focus on manufacturing and retail SMEs with complex product portfolios, high SKU counts, and volatile demand, where forecasting precision directly impacts profitability and resilience.

* Who Part II – The Humans Behind the AI
We are three co-founders with backgrounds in applied AI, software engineering, strategy, and large-scale technology transformation. Together, we combine deep technical expertise with entrepreneurial execution to build the human intelligence behind the AI forecasts.

+ Bonus Track / Recognitions
Winner of WECONOMY 2025/26 by Handelsblatt and UnternehmerTUM, awarded the EXIST Research Transfer program, selected as a Top 10 start-up in the EIT Supernovas Rocket Up program, and invited to present at leading European supply chain and AI events.

GENERIO entwickelt eine skalierbare 3D-Infrastruktur für den E-Commerce. Unsere KI wandelt bestehende Produktfotos automatisiert in produktionsreife, AR-fähige 3D-Modelle um – in Sekunden statt Wochen. Damit lösen wir das zentrale Skalierungsproblem im 3D-Commerce: die wirtschaftliche Erstellung tausender 3D-Assets pro SKU.

Unser Kernprodukt ist eine Integration-ready 3D-API für Enterprise-Kunden. Sie ermöglicht es Handelsunternehmen und Plattformen, ihre gesamten Produktkataloge automatisiert in 3D zu überführen und direkt in bestehende Shopsysteme, AR-Viewer oder digitale Showrooms zu integrieren. Das steigert Conversion-Raten, reduziert Retouren und verbessert operative Effizienz.

Wir verfolgen ein B2B-Enterprise-Modell mit wiederkehrenden Lizenz- und Infrastrukturgebühren. Typische Vertragsvolumina liegen im Bereich von €100.000+ pro Jahr. Ergänzend bieten wir On-Premise-Deployments für IP-sensitive Kunden sowie individuelle Workflow-Integrationen an.

Unsere Wertschöpfung basiert auf einer eigenen KI-Pipeline – kein API-Wrapper. Das sichert Datenhoheit, DSGVO-Konformität und technologische Unabhängigkeit. Ziel ist es, 3D nicht als Marketing-Feature, sondern als skalierbare Infrastruktur für den Spatial Commerce zu etablieren.

GENERIO schafft die technologische Grundlage, um 3D-Produktvisualisierung wirtschaftlich skalierbar und damit zum neuen Standard im E-Commerce zu machen.

inframind develops a sovereign, AI-native operating system for Europe’s energy infrastructure. Distribution system operators and utilities manage hundreds of fragmented data sources across SCADA, GIS, ERP and metering systems. This historically grown IT landscape prevents automation, slows down grid expansion and creates structural inefficiencies. inframind OS integrates, harmonizes and validates these heterogeneous data streams into a synchronized, versioned single source of truth, operated alongside existing systems, not replacing them.
On this unified, versioned data foundation, inframind provides a modular intelligence layer consisting of ontology-driven data modeling, governed AI integration, and application modules for grid intelligence, business analytics, and agent-based automation, enabling automated grid planning, resilience analysis, investment decision support, and end-to-end operational workflows. By reducing up to 90% of manual data reconciliation and coordination effort, inframind significantly accelerates planning and operational processes, ensuring that grid expansion and the energy transition are not delayed by slow, fragmented IT workflows.
Our business model is SaaS combined with forward-deployed engineering: we connect to legacy environments within weeks, implement high-impact use cases, and scale step-by-step across departments. Customers pay annual licenses for the inframind OS and additional modules, plus implementation and customization services.
Digital sovereignty is a core principle. All deployments can run on European cloud infrastructure or on-premise, ensuring data ownership, regulatory compliance and independence from non-European AI companies. inframind provides the data and AI layer that enables utilities to remain in control of their infrastructure while unlocking automation at scale.
By turning fragmented grid data into actionable intelligence, we transform energy infrastructure into a data-driven, automated and future-proof system.

kaarlo addresses one of primary care’s largest unmet needs: psychosomatic (F45) disorders. Up to 25% of GP consultations are driven by psychosomatic complaints, yet 60% of patients remain untreated due to limited time, complex care pathways, and insufficient reimbursement. As a result, patients cycle through repeated consultations and avoidable diagnostics, creating high costs and system inefficiencies.
Our digital business model is based on a scalable, prescription-ready self-therapy platform. After medical diagnosis, patients can immediately start treatment within the kaarlo app. The platform delivers over 30 hours of structured, evidence-based therapy content through more than 65,000 personalized treatment pathways. Interactive, multimedia modules ensure high engagement and adherence. To enhance safety and outcomes, patients can access on-demand psychological support via kaarloDoc.
kaarlo generates revenue through a per-patient reimbursement model, including initial prescriptions and follow-on prescriptions. At an estimated €200 per patient (excl. VAT), our solution costs approximately 10% of comparable outpatient short-term psychotherapy (12 sessions), offering strong economic value for payers and providers.

Ole is the intelligent AI platform for the skilled trades. We create digital tools for the core processes of the energy transition: installations, maintenance, and repairs. Based on real-world expertise and cutting-edge technology, all bundled into a single app. In doing so, we are addressing one of Germany’s last major, structurally under-digitized B2B markets. This is because the skilled trades sector faces a real dilemma: Driven by the energy transition and building renovations, demand is growing, while complexity is increasing significantly due to new product technologies and digital connectivity. At the same time, experiential knowledge is dwindling due to a shortage of skilled workers and demographic change. Digital competitors like Thermondo or 1KOMMA5° are ramping up the pressure. Businesses are losing margins and competitiveness—not because of a lack of orders, but because they lack the right tools. Technology can help, but only if it is combined with genuine expertise and made simple enough to be truly effective on the job site. That is exactly what Ole is: The platform combines deep industry knowledge with modern AI technology to create concrete solutions for everyday work. Ole provides the answer, identifies the replacement part, guides users through maintenance, and triggers the order. This standardizes processes, democratizes knowledge, and makes businesses competitive. The revenue model is two-sided: Ole is offered to the trades as a SaaS model, tiered by company size. Scaling occurs through transaction-based revenue from material procurement via connected wholesale partners or other service providers. The current target market in the DACH region comprises nearly 100,000 HVAC and electrical companies. The IT architecture behind Ole enables both continuous expansion into new digital solutions within the vertical market and rapid adaptation to adjacent trades, thereby allowing the market potential to be gradually increased.

OSPHIM is a digital AI-powered ecosystem that enables plastics manufacturers to optimize production processes through intelligent data analysis and automation.

Our business model is based on a scalable Software-as-a-Service (SaaS) platform that connects directly to existing manufacturing equipment without requiring hardware replacement. OSPHIM combines simulation knowledge, historical production data, and real-time machine signals using transfer learning to generate optimization recommendations even with minimal data availability. This allows manufacturers to improve efficiency (>70%), stabilize quality, and reduce waste (>50%) within hours rather than months. Customers access the OSPHIM App via a web interface, providing real-time process insights, AI-driven parameter optimization, and intuitive decision support for operators and engineers.

Revenue is generated through annual or monthly software licenses per machine. By reducing setup times, scrap rates, energy consumption, and enabling the reliable use of recycled materials, OSPHIM creates measurable economic and environmental value. The platform democratizes access to industrial AI, making advanced manufacturing intelligence accessible not only to large corporations but also to small and medium-sized enterprises worldwide.

Warehouses are under increasing pressure: skilled labor shortages, rising operational costs, growing complexity, and limited real-time transparency of inventory and material flows. Manual forklift operations and blind spots often lead to avoidable accidents, disruptions, and financial losses. At the same time, many facilities still rely on fragmented IT systems and static, non-adaptive control logic, making operations inefficient, reactive, and difficult to scale.
SLAPStack addresses this gap with an AI-powered warehouse operating system, enabling real-time decisions and continuous optimization for both robotic and manually operated forklifts. The platform integrates warehouse management, fleet control, and real-time simulation into one unified system.

VitrofluidiX develops VitroFlow, a simple and versatile organ-on-a-chip (OoC) system that enables the simulation of human organ functions under physiologically relevant conditions. This is achieved by creating a controlled, organ-like environment for human cells directly within the device, without the need for additional laboratory equipment. As a result, cells behave more realistically, enabling more predictive drug testing, improved disease research, and a reduction of animal testing in the long term.

The device is complemented by VitroFlow.Connect, a cloud-based software application that enables automated live imaging and continuous monitoring of experimental parameters via a camera connected to any standard laboratory microscope. The software allows users to remotely observe experiments, record sensor data, and document workflows in a structured and reproducible manner.

VitrofluidiX follows a hybrid business model: the VitroFlow hardware is commercialized through direct sales or leasing, while the digital software component is offered via a subscription-based subscription model that provides ongoing monitoring, documentation, and analysis functionalities. Beyond individual product sales, VitrofluidiX pursues a long-term platform strategy in which VitroFlow serves as a standardized technological foundation for organ-on-a-chip research. Through its open and chip-agnostic design, the system enables the integration of chips, cell models, and modules from external partners. This approach is supported by ongoing collaborations with leading universities, research institutions, and established chip manufacturers, allowing VitrofluidiX to continuously expand the range of compatible models and applications while maintaining a robust and reproducible core system.