Loading…
From PHP to team lead of agents: rethinking judgment, review, and data with Google's Andi Gutmans (Part 1)
Stack OverflowEira May
Summary
Software development is experiencing an architectural shift as artificial intelligence agents transition individual contributors into orchestrators and reviewers of automated workflows. Rather than spending effort on manual coding, developers increasingly focus on defining system architecture, guiding agent execution, and managing operational risk through tiered human and agent review loops. The primary bottleneck for agentic systems lies in preparing organizational data so that agents can effectively reason over enterprise knowledge. To address this challenge, Google published an open knowledge format and is adopting agent-driven semantic ontology generation to infer relationships across messy data estates. This evolution transforms ontology building from manual human toil into automated generation, allowing engineers to focus on high-value data curation and strategic system design.
Context
As software engineering shifts toward autonomous AI agents, organizations face challenges in structuring their enterprise data for agent reasoning and adjusting developer workflows. Manual coding historically consumed the majority of engineering effort, while building semantic ontologies across an entire data estate required prohibitive human labor.
Approach / What changed
Google is transitioning individual contributors into orchestrators of agents who focus on architecture, system design, and risk-managed review loops. To address enterprise data bottlenecks, Google published an open knowledge format and is shifting semantic ontology building from human toil to agent-driven inference via initiatives like Knowledge Catalog, placing humans in high-value curation roles.
Takeaways
- Individual contributors are shifting from direct manual coding to orchestrating AI agents, spending more time on system architecture, operational excellence, and reviewing agent output.
- Deciding between human-in-the-loop, agent-in-the-loop, and agent-on-the-loop workflows is fundamentally a risk-management decision rather than a matter of trust in AI.
- Google is automating semantic ontology creation across enterprise data with initiatives like Knowledge Catalog, moving human engineers from manual ontology generation into high-value curation roles.
Related reading
The new bottleneck
AI coding tools have significantly lowered the cost of generating software, yet many engineering organizations fail to realize overall delivery speed improvements. Applying the Theory of Constraints reveals that eliminating code production as a bottleneck shifts inventory directly into surrounding legacy processes that remain unadjusted. New friction points consistently emerge across underspecified requirements, prolonged design handoff gates, senior engineer review capacity, and external sign-offs from legal or security. Organizations can address these blockers by interrogating legacy agile ceremonies checkpoint by checkpoint to determine if their original constraints still exist. Practical remedies include adopting real-time co-development between product and engineering, treating initial designs as fluid starting points, and restructuring processes around running rapid, high-volume experiments.
Eira MayExplorers, exploiters, and the myth of the 100x engineer