Integration and portability
Data pipelines, APIs and connectors, maintained as one codebase, with data and configuration export in documented formats so that a model or hosting provider can be replaced.
Capabilities
What Artabel builds, and where it has been used.
Applied AI and the engineering around it: the sectors we work in, the systems we build and how they are delivered. How they are evaluated and governed is set out under Research and evaluation.
Sectors
Sectors we work in.
MedTech
Medical AI
Imaging and multimodal interpretation, clinical decision support, structured documentation, research workflows, dataset curation and quality auditing, and evaluation with clinical reviewers against agreed criteria.
Where a system could qualify as a regulated medical device, that question is settled during scoping, before the design depends on the answer.
Industry
Industrial AI and operations
Predictive modelling and maintenance, anomaly detection, document intelligence, workflow automation, systems integration, instrumentation in hostile environments and real-time factory safety monitoring. Delivered into aluminium production, metallurgy and semiconductor manufacturing.
Agri-food
Agriculture and agri-food
Decision support for farms and agri-food producers: planning around weather, treatments, irrigation and field work from forecasts and operational records, with the reasoning behind each recommendation shown to the person who decides.
Artabel’s own product, Satoria, piloted with a specialty-crop producer in France.
Heritage
Culture, heritage and public knowledge
Source-grounded interpretation for museums and archives: ingestion of heterogeneous and partly analogue collections, rights-aware access, provenance and attribution, multilingual delivery, and interfaces the public uses unsupervised. The engineering is the same retrieval and document work as elsewhere; what differs is that the rights position changes record by record.
What we build
What we engineer.
Ten areas of capability.
Enterprise
Enterprise information systems
Configurable systems for production, procurement, logistics, accounting and document approval, with role-based access, reporting and integration with the systems already in place, maintained over many years.
Delivered: VRPIS, the enterprise platform of an aluminium producer, in daily use since 2017.
Retrieval
Source-grounded retrieval and RAG
Ingestion, OCR, layout and metadata extraction across heterogeneous archives; hybrid lexical and semantic retrieval; generation that cites the passage it used. An unsupported assertion in generated output is treated as a defect.
Built and running: an eight-year corporate document archive with AI classification and analytics; an ongoing intelligent-document programme since 2023.
Vision
Computer vision and medical imaging
Classification, detection and segmentation on images that are noisy, scarce or clinically sensitive, with anatomical mapping, area quantification and observer-independent scoring.
Research reference: automated quantification of eczema lesions by anatomical region — fingertips, fingers, palms, dorsum, wrists — with the University Hospital of Basel.
Language
Speech, language and structured extraction
Speech recognition and structured documentation from professional dictation and consultation audio; entity and relation extraction from unstructured text into schemas your systems already use.
Research reference: automated transcription and summarisation of dermatology consultations, producing structured clinical documentation with human validation — with the University Hospital of Basel.
Prediction
Predictive modelling and time series
Forecasting, anomaly detection and remaining-life estimation from sensor and process data, evaluated against the cost of a false alarm and of a missed failure.
Delivered for onsemi, a semiconductor manufacturer: aggregation of more than fifty sensor streams into AI-driven lifetime prediction for electrical parts.
Data-centric
Data-centric AI and dataset quality
Auditing a dataset before modelling it: off-topic samples, near-duplicates, label errors and the quality problems that silently cap what any model can achieve.
Research reference: automated cleaning of large image collections — detecting off-topic samples, near-duplicates and label errors.
Synthetic
Synthetic and generative data
Generation of training and test material where real data cannot be shared, with metrics for fidelity, diversity, downstream utility and privacy leakage assessed separately.
Research reference: a project on synthetic-data benchmarking, developing evaluation metrics and comparative benchmarks.
Agents
Assistive agents under human authority
Task-scoped assistants inside a defined professional workflow: bounded tool access, explicit source attribution, a named human who approves before anything is released, and a manual fallback.
Research reference: automation of digital-forensics workflows, where each step is attributable and a qualified examiner approves the result.
Interfaces
3D, VR and immersive interfaces
Configurable 2D and 3D visualisation and immersive environments for teaching, research and technical inspection.
Edge & IoT
Smart devices, sensing and IoT
Instrumentation for hostile environments, real-time position and condition monitoring, and the integration that puts readings into the system of record rather than onto paper.
Delivered for Slovalco: automated molten-metal level measurement, and a vehicle beacon for real-time factory safety monitoring.
Delivery
Delivery and operations.
Artabel has been putting software into other organisations’ existing systems since 2004. The AI work builds on that integration and maintenance practice.
Environments
Company-operated development, integration, test and delivery environments, with version-controlled repositories and separated test workflows. Compute and storage in Switzerland or the EU for routine AI workloads.
Monitoring and MLOps
Dataset and model version control, task-level evaluation harnesses, error analysis, human-review queues, structured logging and deployment monitoring to track performance and investigate drift.
Handover
Documentation, automated tests, a support route and the cost of running the system, so the client can operate and change it without us.
How to start
Four ways to begin.
Feasibility sprint
Two to six weeks. One question, your data, and a written answer: what is achievable, what it would cost, and where it is likely to fail.
Build and validate
A working system plus the evaluation that shows whether it beats the current way of working, delivered as one package.
Independent evaluation
For systems built by others: benchmark design, the evaluation protocol and deployment verification as a separate engagement, as set out under evaluation.
Research consortium
Artabel joins European research projects as the engineering partner. Roles and participation details are on the Programmes page.
Next step
Tell us what the task involves.
Describe the task and the constraints around it: where the data sits, the quality the result must reach, and the deadline.