The focus in AI engineering has shifted from the agent itself to the surrounding system or 'harness.' This includes managing workflows, context, permissions, and tools. Engineering these reliable systems is now seen as more critical for delivering value than simply prompting a more powerful model.
A new 'loop engineering' paradigm structures work into two parts: an 'inner loop' for autonomous AI execution and a human-managed 'outer loop' for strategic direction and oversight. This model clarifies the division of labor, ensuring humans retain control over key decisions while leveraging AI for execution.
Rather than passively waiting for model improvements, 'skill engineering' is emerging as a discipline. It involves actively encoding expert workflows, quality gates, and even subjective 'taste' into portable components for AI agents, allowing organizations to consistently improve agent performance on specific tasks.
The initial excitement for fully autonomous agents has cooled. The industry now recognizes that 'autonomy without structure creates as much slop as leverage.' The new focus is on building systems where humans are central, providing direction, oversight, and strategic decision-making to guide agentic work.
To control costs, security, and governance, enterprises are moving from interactive, ad-hoc agent use to a 'software factory' model. This approach systematizes the entire work lifecycle, automating processes and minimizing the risks associated with inconsistent human operation of powerful AI tools.
The primary interface for knowledge work is shifting away from traditional applications like IDEs. Professionals are increasingly initiating and managing complex tasks within conversational, agent-driven environments (e.g., in Slack with Claude tag), signaling a fundamental change in human-computer interaction.