From Tab Completion to 100% Code Written by Agents
- 2 minsFrom tab completion to 100% code written by agents.
That escalated quickly.
I still remember the first time I used GitHub Copilot in 2021. You typed a comment, hit tab, and half the function appeared. It felt futuristic.
Fast forward a few years.
Now we are talking about workflows where agents plan, write, test, refactor, and iterate on code with minimal supervision. Not suggestions. Not snippets. Entire task ownership.
Recently, Cursor cofounder Michael Truell shared data showing how agent usage has overtaken simple tab completion in product usage metrics.
That chart is telling.
Agent requests per tab accept rising through 2025 and crossing 1x by Jan 2026 — the point where agent workflows overtake tab completion.
It shows agent requests per tab accept climbing steadily through 2025, eventually crossing 1x. In other words, developers are triggering full agent workflows more often than accepting inline completions.
That is a profound behavioural shift.
Tab completion augments the developer. Agents assume responsibility for chunks of the workflow.
This is not just a UI evolution. It is a transfer of execution.
If agents become the default way code is produced, several things follow:
-
Architecture matters more than syntax. You are no longer optimising keystrokes. You are designing systems agents can operate safely within.
-
Control becomes strategic. Who reviews? Who authorises? What guardrails exist? When an agent can touch multiple files and refactor broadly, governance is not optional.
-
Headcount is no longer a clean proxy for output. If agents handle more routine engineering surface area, the leverage per developer increases significantly.
We are moving from human-written code with AI assistance to human-supervised systems with AI execution.
That is a different era.
Copilot showed us that AI could complete our thoughts. Agent workflows are starting to complete our tasks.
The real leadership question is not whether this works. The data suggests it already does.
The question is how much execution authority we are prepared to delegate, and how we intend to control it.
The shift from tab completion to agent default may look incremental on a chart.
It is not incremental inside the organisation.
Originally posted on LinkedIn: https://www.linkedin.com/posts/davidlowjw_from-tab-completion-to-100-code-written-share-7434219643118034944-Lo-9