Today's Agenda
Hello fellow humans! Today, there is big news around agents, and for most people, it may not be clear what that means, but the AI powerhouse companies are trying to reshape our entire information ecosystem, and the pace of change is breathtaking. OpenAI and Nvidia have both declared that the 2025 is the year of the agent.
News
Learning, Adaptation, and Skills Development in the AI Era
AI is transforming learning and skill development, both for AI systems themselves and for humans using AI. The emphasis is on adaptive learning systems, personalized education, and the need for continuous upskilling to remain relevant in an AI-augmented workplace. Anthropic's recent work on "Agent Skills" offers a new blueprint for building AI systems that can learn and adapt more effectively.
Must-Have Agentic AI Skills for Product Managers
For product managers in the job market, the essential skills for working with agentic AI systems include:
AI & ML literacy
Context and data management
Workflow orchestration
Ethical oversight
Product managers need to understand how to integrate autonomous AI agents into products while maintaining human control and ensuring ethical implementation.
The Teacher is the New Engineer: AI Enablement
Generative AI is probabilistic and adaptive; it learns — both intentionally and unintentionally — from interaction and operating in gray zones of uncertainty. We are shifting from traditional software engineering to AI enablement, where the focus moves from deterministic programming to teaching and guiding AI systems through continuous interaction and feedback.
This isn’t reinforcement learning, a practice that Andrej Karpathy thinks is a dead-end for AI, but rather establishing human practices like:
Onboarding the AI systems
Context engineering
Prompting
Human-in-the-loop review practices
Strong critical thinking
AI can be an accelerator, but it requires real human skills to produce valuable work. The more explicitly and thoughtfully we can instruct AI to work for us, the more it will be able to deliver what we need.
Architectural Decisions: A Human-Led, AI-Powered Approach
Advocates for "human in the loop" (HITL) approaches in AI-augmented decision-making systems. Emphasizes the critical need for human intervention capabilities in automated systems to prevent errors and maintain accountability. Discusses architectural patterns that preserve human agency while leveraging AI capabilities for enhanced decision-making.
Prioritization Starts With Strategic Prioritization
John Cutler is a longtime leader in the product thinking space, who talks about everything from how to understand complex data to how to structure organizations to deliver great products. I share this essay because one of the most important jobs in using AI effectively is setting strategy and priorities. He offers strategic frameworks and walks through:
The four benefit types of a strategy
Drivers that influence those benefits
Helmer’s 7 powers and other strategic frameworks
The lifecycle of strategic advantages
And after all that, then you can prioritize!
Radical Candor
If you’re using [AI] just for efficiency, I think you’re under-utilizing it. That is our big learning from observing people who use it really great versus folks who have AI access, who have CoPilot, but they're using it still in… ways that are incremental.