Your AI applications and agents, from automation to customer experience

Reviewing tens of thousands of CEE files by hand was holding back the growth of the business. We built Athos, an AI-first platform based on a system of agents combining OCR and LLMs: it classifies files, checks their regulatory compliance and isolates anomalies for the reviewers. The result: a bottleneck turned into a high-volume, streamlined flow, with human effort refocused where it adds value.

Mercialys faced many time-consuming operational processes and wanted to define its AI strategy. We scoped the whole approach from start to finish: setting up a steering committee, running a roadmap of projects prioritised by ROI, then rolling out a first four agentic solutions in six months – contract analysis, an internal chatbot, investment memo generation and automated market monitoring. A concrete, measurable transformation, rooted in how the teams actually work.

Reading and summarising complex tenders by hand took many hours of work per bid. We built a bespoke AI agent that extracts technical and regulatory requirements from documents running to several hundred pages, and flags inconsistencies automatically. Analysis time has been cut drastically.

Finding a specific clause or contract among more than a million documents was a daily challenge for Gide's 500 lawyers. We built and deployed a generative AI platform that searches the entire corpus and automatically extracts the key metadata from each document. What used to mean tedious manual digging is now a query that takes a few seconds, with precise, contextualised results.

Training You prepares students for interviews at investment banks and consulting firms, an oral exercise that's hard to practise alone. Galadrim built the voice training app integrated into its course platform: the AI interviewer asks its question out loud, the candidate answers aloud, and a language model scores the answer and follows up on weak points. An internal ranking adjusts the difficulty as the interview goes on, and the candidate gets an audio debrief, question by question.

Urgo Medical equips 160,000 healthcare professionals with Healico, its wound-monitoring app, and wanted to put its specialists' expertise within reach of non-specialist nurses. Galadrim built the conversational assistant on a closed corpus of validated medical references: it queries the indexed literature, cross-references the condition with the state of the wound bed to suggest a protocol, and always cites the original scientific source. Guardrails rule out any answer without a source or that strays into diagnosis, which keeps the product outside the scope of medical device regulation.

Commercia, a French preparatory school for business schools, wanted to speed up its ECG-track students' progress in maths – a subject largely worked on alone, where the teacher's feedback arrives after the fact. Galadrim built MathexIA, the app that assesses each student's level, generates worksheets and quizzes on demand calibrated to their gaps, then marks their work. Each student follows a path built for them, and the teacher starts each class on the real sticking points of the group.