IT4LIA’s Horizontal Services guide users through every stage of AI development— from the first experiment to trusted deployment. They provide the environments, tools, and expert support needed to develop, test, and scale AI workloads on high-performance computing infrastructure. By combining human expertise and cutting-edge technology, they help users design efficient and trustworthy AI systems, validate their performance, and ensure transparency and regulatory compliance. Across all domains, these services accelerate innovation, strengthen collaboration, and transform ideas into impactful, real-world AI solutions.
Discover Horizontal Services
Guiding your AI journey from first experiments to trusted, scalable solutions on HPC
Application examples
AI SETUP
Reducing the “time-to-first-run” is critical for research productivity.
This service facilitates, among other activities:
- Hugging Face Integration: Loading pre-optimized pipelines directly onto the HPC cluster without the need to build custom environments from scratch.
- Autonomous Job Submission: Providing end-to-end guidance on login protocols and best practices for submitting scalable AI workloads.
- Dependency Management: Offering pre-validated software stacks where all library dependencies are already configured for maximum hardware efficiency.
Case Study: The service provides comprehensive onboarding for a first-time AI Factory user. By guiding them through login configurations and job submission best practices, it enables them to quickly become autonomous on the infrastructure so they focus entirely on their AI development work.
AI DEVELOPMENT
Organizations with vast documentation require efficient ways to retrieve and synthesize information.
This service facilitates, among other activities:
- Customer Care Automation: Indexing FAQs and technical manuals to allow support pipelines to automatically resolve complex tickets based on precise customer history.
- Internal Policy Access: Deploying chatbots that offer employees instant, natural language access to company policies and procedures via Vectoria.
- Technical Log Synthesis: Implementing high-performance retrieval across technical logs and manuals using specialized tools like Haystack and FAISS.
Case Study: The service supports a telecom leader in indexing technical manuals and support logs, creating an automated pipeline that delivers precise solutions to technical staff, significantly reducing manual search time.
AI DEVELOPMENT
Organizations often require AI models tailored to their specific data and operational contexts.
This service facilitates, among other activities:
- Time-Series Modelling: Developing forecasting and anomaly detection models for sensor and machine data streams.
- Computer Vision Applications: Developing object detection and quality control models for industrial inspection scenarios.
- Framework Selection: Supporting implementation using frameworks such as PyTorch, TensorFlow, scikit-learn, XGBoost, YOLO, OpenCV, and ONNX Runtime.
Case Study: The service supports an enterprise in developing AI models for automated visual inspection and predictive maintenance using domain-specific datasets and workflows.
AI TEST
Bridging the gap between a local prototype and high-traffic production environments is critical.
This service facilitates, among other activities:
- Resource Estimation: Helping companies pre-training or fine-tuning models on custom data to accurately estimate required GPU hours and infrastructure scaling.
- Inference Dimensioning: Profiling workloads for vLLM-based endpoints to determine the required GPU setup, parallelism, and batching strategies.
- Load Analysis: Identifying expected latency and throughput under load to ensure service stability for citizen-facing or enterprise applications.
Case Study: A startup builds a citizen-facing chatbot on an open-source LLM. The service profiles the workload and identifies the necessary GPU setup and batching strategy, which ensures the startup supports production traffic levels with optimized latency.
AI TRUST
Organizations deploying AI systems often require objective evidence regarding trustworthiness and operational behaviour.
This service facilitates, among other activities:
- Security and Robustness Testing: Assessing vulnerabilities, abuse scenarios, and system robustness.
- Observability and Control Assessment: Evaluating model transparency, observability, and controllability.
- Guardrail Validation: Testing safety mechanisms and operational constraints.
Case Study: The service supports a telecommunications provider in comparing two LLM-based solutions through sandbox-based testing and standardized trustworthiness assessments prior to deployment.
AI TRUST
Sensitive data applications require rigorous adherence to privacy laws.
This service facilitates, among other activities:
- Gap Identification: Utilizing specialized self-assessment tools and in-depth interviews to identify critical technical and legal compliance gaps.
- Data Privacy Engineering: Implementing robust anonymization or pseudonymization strategies for training datasets to comply with GDPR.
- Risk Mitigation: Deploying privacy-preserving machine learning techniques to minimize patient re-identification risks in clinical environments.
Case Study: A technology provider develops an AI system for cardiac diagnosis using sensitive medical imaging. The service conducts technical analyses to ensure alignment with data protection frameworks and provides a bespoke compliance roadmap that enables secure market deployment.