Saudi Ministry of Industry.
An AI strategy for the Ministry of Industry's Seneai platform (Saudi Vision 2030) — around 60 potential use cases assessed on impact and feasibility, distilled to 7 prioritised initiatives and a phased roadmap, plus a working proof of concept for AI-assisted factory inspections.
- Industry
- Public sector / AI strategy + PoC
- Timeline
- Assessment + roadmap + PoC
- Outcome
- ~60 use cases → 7 priorities + a working PoC
The problem
Saudi Arabia's Ministry of Industry and Mineral Resources runs Seneai — the platform industrial investors use to apply for licences, manage permits and stay compliant, and a flagship piece of the Vision 2030 digital agenda. That AI could help was obvious; where exactly, and whether it was actually feasible, was not. The Ministry needed a clear-eyed answer grounded in its real processes and its data — not a generic list of AI ideas.
What I built
A strategic assessment that turned a broad ambition into a concrete, sequenced plan — and a working proof of concept to show it was real.
- From ~60 ideas to 7 priorities — drawing on stakeholder interviews, the platform's requirements and a review of what was technically possible inside the Kingdom, I assessed around 60 potential AI use cases against one framework: business impact (value to investors and to the Ministry) versus feasibility (data readiness, legal fit, technical complexity). Seven rose to the top.
- Grouped around what matters — the seven cluster into three themes: better cross-language service for investors (an Arabic–English document translator, a bilingual query assistant), operational excellence (a risk-scoring engine, a document-completeness validator, a licence-type recommender, a processing-time predictor), and inter-agency coordination (a single cross-ministry status tracker).
- A phased roadmap, not a wish list — sequenced into quick wins (0–3 months), strategic enhancements (3–6 months) and a longer transformation (6–12 months), each with its data, infrastructure and stakeholder prerequisites spelled out.
- A working PoC for field inspections — I built a proof of concept for an AI assistant that reads inspectors' factory photos, flags equipment, safety and compliance issues, and drafts a structured inspection report in Arabic or English for the inspector to review and approve. Specialised AI agents handle different aspects — equipment, safety — and a lead agent merges them into one report. It was designed to run inside a cloud region in the Kingdom to meet data-sovereignty rules, with a person always in the loop on the final report.
The outcome
A prioritised portfolio of AI initiatives with a phased implementation roadmap — and a working proof of concept showing the highest-value idea was buildable, not just plausible. The Ministry got a plan it could act on: what to do first, what it would take, and what the result looks like in practice.