Skip to main content
đŸ€– Running agents for a team, not just yourself?Get an independent review of identity, secrets, failover, observability and governance.Assess your agent platform
Panel discussion stage at PlatformCon London 2026
Platform Engineering

Kief Morris on AI Agents and Being 'Human on the Loop'

Kief Morris of Thoughtworks on what engineering identity means when agents write the software: build the system that builds the software, not just the software.

Luca Berton

· The Production AI Expert

Published ·Updated ·4 min read

What does it mean to be an engineer when agents write the software? That was the question behind my short conversation with Kief Morris (Thoughtworks) on the floor at PlatformCon Live Day London 2026, and his answer reframes the entire debate about AI’s role in engineering more usefully than most of the takes I heard that day.

Agents Are the Topic, Not the Problem

Kief’s opening observation was blunt: agents are the defining topic in the industry right now, but the real challenge was never whether to use them — it is learning to use them well. That distinction matters more than it sounds. Most of the anxiety around agentic development comes from treating “should we adopt agents” as the open question, when the harder and more useful question is “what does competent use actually look like.”

Human on the Loop, Not Human in the Loop

Kief’s article on MartinFowler.com, “Human on the Loop,” reframes the engineer’s role in a way that survives contact with how agentic development actually works day to day: the job is to build the system that builds the software, not just the software itself.

The distinction between “in the loop” and “on the loop” is doing real work here. “In the loop” implies a human checking every individual output — a model that does not scale and quietly degrades into rubber-stamping once the volume gets high enough. “On the loop” describes someone who has built and continuously tunes the system — the guardrails, the evaluation criteria, the escalation paths — that determines whether the agent’s output is trustworthy in the first place. The engineer’s attention moves up a level, from individual outputs to the system that shapes all of them.

The Talk: “Humans on the Loop”

The same afternoon Kief took the stage with a session titled “Humans on the Loop: Taking Agentic Engineering End to End”. A slide with the session title lists him as Distinguished Engineer, Technology Advisory Services at Thoughtworks, with the cover of his O’Reilly book Infrastructure as Code alongside.

Session-title slide “Humans on the Loop: Taking Agentic Engineering End to End” by Kief Morris, Distinguished Engineer, Technology Advisory Services, Thoughtworks, at PlatformCon London 2026

The session-title slide from Kief Morris’s PlatformCon Live Day London talk.

The slides I photographed move through the argument in steps, beginning with “vibe coding (basically)”: an idea becomes a prompt, the agent builds, working software comes out, and a human reviews it.

Slide “Vibe coding (basically)” showing a loop from idea to prompt, build, working software and review

“Vibe coding (basically)”: prompt, build, working software, review.

The next slide adds a spec at the start and tests at the end: “Spec-driven development with tests” puts the spec and the tests around the build step, so the review loop has something concrete to check against.

Slide “Spec-driven development with tests” showing a loop from idea to spec, build, working software and tests, with a review step

“Spec-driven development with tests”: the same loop with a spec in front and tests behind.

The ideas also connect with the “agentic coding harness” slide from the same deck, defined as “a system that drives the LLM to make good code”. That is the “system that builds the software” in practice.

Sensing Deliverability and Operability

The slides also cover how a human on the loop knows whether the output is any good. One slide asks “How do we sense deliverability?”, with DORA metrics as the lagging sensors and mutation testing and fuzz testing as leading sensors.

Slide “How do we sense deliverability?” listing lagging sensors: DORA and things like that

“How do we sense deliverability?” with DORA as the lagging sensor.

A follow-up slide asks the same for operability and splits the answer into lagging sensors (DORA deliverability metrics, user satisfaction, business operations metrics such as transactions and cost, commercial metrics) and leading sensors (deployability, performance and scalability testing, failure scenario testing, penetration testing, compliance auditing).

Slide “How do we sense operability?” listing lagging sensors and leading sensors

“How do we sense operability?”: lagging sensors on the left, leading sensors on the right.

My take: this lines up with the “what does not change” point from our conversation. Quality and the path to production still need measuring, and the leading-sensor list reads like the checks a platform team can automate in the pipeline.

What Does Not Change

The part of the conversation I keep coming back to: managing quality and the path to production remain paramount, exactly as they always have. AI changes the tooling — what generates the first draft of the code, what catches an obvious mistake, what accelerates a refactor — but it does not change the fundamentals of what makes software fit to ship. A codebase with weak test coverage, unclear ownership, or no deployment discipline does not get healthier because an agent wrote more of it faster.

This is the same conclusion I keep arriving at from a different angle in production guardrails for AI agents: the discipline lives in the system around the model, not in the model itself.

Find His Work

Kief pointed me to infrastructure-as-code.com for the deeper material — or, as he put it, just search “Kief” (K-I-E-F) and the right results surface. Given how much of the current agentic-AI conversation is repeating lessons Infrastructure as Code already learned about treating operational systems as engineered artifacts, that back catalog is worth the search.

About the Author

I am Luca Berton, AI and Cloud Advisor. I work at the intersection of platform engineering, cloud security, and enterprise AI deployments. Book a consultation.

Free 30-min Production AI consultation

Book Now