Practitioner, not platform.
I am an independent AI and data systems consultant with 10+ years inside high-trust, high-scale organizations: the Government of Canada, the United Nations, Lyft, Oliver Wyman and major enterprises across North America and the Middle East. I have shipped production systems in every one of those environments.
The advantage of working with me is that the senior strategic thinking and the hands-on execution come from the same person. I can define the architecture with your executives, then personally build the system with your teams, and stay accountable through to operation.
Complex organizations, real production systems.
Government of Canada
Five years as a senior data developer with a federal digital service: cloud data platforms on AWS, orchestration, infrastructure as code, governed self-service analytics, and delivery inside real public-sector governance and stakeholder constraints.
United Nations
AI engineer for a multilateral organization across Geneva and New York: enterprise RAG on Databricks, LangGraph agentic workflows, MCP integration, and evaluation pipelines serving teams in many countries.
Lyft
Production machine learning at consumer-technology scale: computer vision shipped into a live operational platform, validated against human benchmarks with controlled experiments before rollout.
Oliver Wyman
Senior consultant based in Dubai with the global strategy firm: petabyte-scale network analytics and investment optimization for major Gulf telecom operators, with results presented to CEOs and senior leadership.
Yannick Bounouar
M.Sc. Information Systems, WU Wien. B.Sc. Mathematics, Université de Montréal. Systems design and quantitative foundations, applied for a decade to production AI and data platforms.
The through-line of that decade: taking complex AI and data initiatives from ambiguity to reliable production. Ingest, lakehouse, retrieval, agent tooling, evaluation, delivered with governance, lineage and observability so the system remains operable after handoff.
- Ship small, observe, iterate. Prototype against real data and real access controls; eval is the gate to scale, not a slide.
- Governance before scale. Lineage, auth and provenance are part of the build, not a later ticket. If it cannot be traced, it does not ship.
- Hand to the team that operates. Docs, runbooks and ownership land before launch. The system must run without me.
Yannick Bounouar: M.Sc. Information Systems, WU Wien; B.Sc. Mathematics, Université de Montréal. Independent practitioner in production AI and data systems.
Discuss your AI initiative
30 minutes to talk through your initiative, the data behind it, and the shortest path to a production outcome. Direct booking, no form required.
Prefer email? info@yannickbounouar.com. We review promptly and propose a next step.