Higher quality
Edge cases, confusing copy, incorrect helper text, and off padding are no longer acceptable misses.
October 2026
Expectations &
productivity in the
agentic era
Expectations. Adoption. Productivity?
⚠ This talk has a two-month shelf life.
01 · Expectations
Pre-AIHigh-quality outcomes.
High-quality outcomes.
High ownership.
→
NowHigher quality.
Higher quality.
Larger scope.
Larger scope
Engineers must take on larger scopes of work and own more of the outcome.
Faster pace
Engineers must move at a much faster pace while maintaining quality.
04Apply judgment.
Judgment can no longer be centralized. Every engineer must be able to apply more judgment than ever before.
A note on measurement
Adoption metrics inform
the conversation. They are
not the measure of success.
token usagelines of codetickets closedagents running in parallelbeing on the bleeding edgebusy work
What matters is the product value we create today and our ability to create more value tomorrow.
“When a measure becomes a target, it ceases to be a good measure.”
02 · AI adoption?
Without AI, meeting the new bar
would be virtually impossible.
AIFaster, less error-prone, and no longer specialized
HumanWhere judgment is required
03 · Productivity
AI is better.
Our assumption at Renterra.
01Enforcing the baseline
Skills and documentation encode the standard.
02Implementation
Solving bounded problems with speed and consistency. (Shout out to Matt Pocock’s improved codebase architecture.)
03Code review
AI approval. Deep reviews that scale across every change.
04Planning
Domain impact and edge case analysis.
05UI and design
Faster, less error-prone, and no longer specialized.
03 · Productivity
The job of engineering
For us at Renterra, it focuses on:
01Improving the harness
Making it easier to build well. Don’t fight the models.
02UX
Responsible for how it feels to use.
03Ownership and communication
Owning the outcome and keeping the team aligned. No slop grenades and no meat proxies.
04Applying judgment
Knowing when to stop, when to go, and whether you’re on the right path.Slowing down at decision time is often faster in the long run.
How well are engineers applying judgment?
Open questions?
What motivates engineers?
What happens when implementation is no longer the hard part?
Applying judgment?
Is there a better way to measure this? How do we coach for this?