Speaker
Abstract
Coding agents produce 741% more code, and only 30% more software ships as a result. The gap is human review: people top out around 400 lines an hour and agents are not slowing down. This talk is for engineering leaders trying to decide what to do about that. It covers what happens to code review when the reviewer is the bottleneck, which review work is moving to agents and automated verifiers and which is not, what the change does to hiring and to the junior pipeline, and where the "product engineer" role is landing as the line between writing code and shipping software dissolves. It draws on published data from the largest practitioners in the space. Leaders leave with a framework for deciding what their people should read, what they should delegate, and how to measure whether the new arrangement is working.
Interview
Everybody has started using LLM assistance to generate code, and that has eliminated the speed of coding as a bottleneck, but revealed a new one: the speed of reviewing the code we produce, to make sure it's correct and safe. My talk is about the current state of the art in accelerated code review.
Data shows that coders using LLM assistance wrote more than 700% more code but released only 30% more actual software; that gap is all about the bottleneck of review.
The whole space is brand new. Nobody's really sure what works and what doesn't. Can we automate review? Can we eliminate review? Can we avoid review with better specs? People are trying all sorts of strategies.
That while this is an unsolved problem, there are better strategies out there than just "review faster", which demonstrably doesn't work.
Topics
$2,995, Conference (3 days). Current pricing ends November 10th. All pass options.
QCon San Francisco 2026 is a three day conference for senior software engineers, architects and team leads. An international program committee of working engineers selects every session. Patterns and practices, not products and pitches.
From the same track
Wednesday 18 November
10:35 Seacliff ABC Session Builder-Driven Development: Rebuilding the Organization Around the Builder Jyothi Nookula Product Leader Director with 13+ Years Driving AI Product & Platform Innovation, Previously @Meta, @Amazon, and @Etsy Most teams still run agent development the way they run traditional software projects. Requirements are written up front, builders execute against them, and a demo closes out the cycle. That model works poorly for agents, because the most useful information rarely comes from a planning document. 11:45 Seacliff ABC Session Staff Plus Engineering Legacy as Leverage: How Brownfield Work Builds Technical Judgment Shine Garg Founder of Uncharted Path Breakthrough, Former Staff Software Engineer, 15+ Years Building and Scaling Backend, Data, Machine Learning, and Platform Systems Legacy systems are often treated as a necessary evil: work to be endured before moving on to more “real, strategic” projects. 13:35 Seacliff ABC Session AI/ML Product Engineering: When code costs nothing, what's next? Laurie Voss Head of Developer Relations @Arize AI, Co-Founder of npm, Inc., Developer for 30+ Years Coding agents produce 741% more code, and only 30% more software ships as a result. The gap is human review: people top out around 400 lines an hour and agents are not slowing down. This talk is for engineering leaders trying to decide what to do about that. 14:45 Seacliff ABC Session AI You Can't Read the Code Anymore - Verifying Software in the Age of AI Omer van Kloeten Principal Engineer @Forter, 25+ Years in Software Engineering, Now Building an AI-Agent Framework to Tame Legacy Systems and Reduce KTLO AI writes most of the code now, so the hard part of engineering has moved from authoring to the part we’re worse at - verifying. With the incentives all pushing toward shipping faster and faster - the rational move is to skim the diff, click through a little manual QA, and ship on vibes. 15:55 Seacliff ABC Session When Everyone Can Code, Who Owns Production? Sanjay Singh Staff Software Engineer @LinkedIn Specializing in Large-Scale Distributed Systems, Traffic Infrastructure, Reliability, and Zero Trust Security AI is making it dramatically easier to create and change software, but writing code is only one part of the software lifecycle. Every change still needs to be deployed, observed, operated, diagnosed, and kept reliable in production.