AI, IT & SaaS categories

Models and inference infrastructure that operate under a specific jurisdiction’s law, on data that cannot leave the country. Covers national and sector-specific model programmes, in-country and region-pinned inference, and the governance wrapper that makes them auditable.

The leading commercial frontier models and the platforms built on them, brought to market in the region with local contracting, in-country billing and enterprise-grade controls.

Security for AI and security using AI — model and agent security, data-loss prevention across AI surfaces, threat detection and response, and the assurance work regulated buyers require before anything goes live.

Developer-facing AI — code generation, review and migration tooling — alongside open-weight models that can be self-hosted, fine-tuned and run at materially lower inference cost.

The connective layer between models and systems of record — API management, event streaming, retrieval pipelines and the integration work that turns a model into part of a working process.

Running AI as a managed cost centre and a managed pipeline — token and GPU spend visibility, allocation and forecasting, plus the build, evaluation and deployment machinery behind model and agent releases.

Agent lifecycle management, multi-model routing, policy enforcement, evaluation and audit trails — the control plane that lets an organisation run many agents without losing track of what they are doing.

Cataloguing, lineage, quality, classification and residency controls. Most stalled AI programmes in this region are data problems wearing an AI costume; this category is where they get resolved.

Process mining, change management, adoption measurement and value tracking — the discipline that connects a deployment to a number someone in finance recognises.