Gen AI: What tools and resources do you find indispensable for enhancing your capabilities?

My thought:

At this stage, I approach Generative AI less from a specific toolset perspective and more from an enablement and governance standpoint. Before tools become indispensable, clarity around data sources, data quality, and usage boundaries is essential.

From a project and portfolio management view, the most valuable resources today are not individual AI applications, but frameworks that help teams assess data readiness, define acceptable use, and ensure transparency and accountability in AI-supported outputs. Traditional tools for data validation, documentation control, and visualization remain critical, with GenAI acting as an augmentation layer rather than a replacement.

As adoption matures, I see success coming from integrating GenAI thoughtfully into existing workflows supported by strong data management practices and clear human oversight rather than relying on tools alone.

Our approach focuses on a practical GenAI stack:

• Data sourcing: curated internal data and targeted synthetic data where gaps exist

• Data prep & quality: SQL, Python, and Power Query for cleaning, validation, and lineage

• Model use: base LLMs with prompt engineering and RAG before any fine-tuning

• Deployment: API-based integration with cloud platforms and CI/CD controls

• Governance: access control, PII redaction, human-in-the-loop reviews

This keeps models accurate, secure, and easy to evolve as data changes.

Good data and constant testing.

What is the most significant barrier you face as a PM, in uising GenAI for more than just administrative simplification?

Developing a culture, the adopt the usage of Gen AI in the Industry require the technological development of the society where the industry is located, there business interest, exposure and knowledge of AI and the management value for AI. Africa is at the bottom of its usage, and a lot has to be done.