tech
January 29, 2026
Rob Valk
Rob Valk, Engineering CTO at Deloitte Australia, explores the future of software engineering for AI-fuelled enterprises.
TL;DR
- Clients are more concerned with AI adoption, safety, compliance, and data privacy than technical AI capabilities.
- Current frontier AI models succeed only 50% of the time on tasks, highlighting the need for checks and balances.
- AI capabilities are doubling every seven months, requiring organizations to recalibrate ROI and explore new use cases.
- Software engineers are transitioning from writing code to managing AI systems, emphasizing architecture and management skills.
- Agentic AI systems require human judgment for scaling and productionizing, focusing on compliance, reliability, safety, and explainability.
- Deloitte's 'human-led, tech-enabled transformation' approach prioritizes human outcomes over technology itself.
- Human judgment is essential for driving AI adoption, ensuring AI serves as a tool to enhance human creativity and productivity.