When AI writes a large share of code and agents take on whole tasks, the shape of engineering work changes. The hardest questions for a CTO become organizational.
What changes in the work
- More specifying and reviewing, less typing. Clear problem framing, task decomposition and review judgment become core skills.
- Platform matters more. Gateways, evals, agent pipelines and fast CI are shared assets that multiply every team.
- Smaller teams can own more — if review, testing and operations keep up.
Signals from the market
- In April 2025 Shopify CEO Tobi Lütke published an internal memo stating that "reflexive AI usage is now a baseline expectation at Shopify", and that before asking for more headcount and resources, teams must demonstrate why they cannot get what they want done using AI.
- Stanford Digital Economy Lab researchers (Brynjolfsson, Chandar and Chen), using ADP payroll data, report that employment of young workers aged 22–25 in AI-exposed occupations now stands 19% below where it would be had it kept pace with less-exposed peers (revised August 2026). They and others continue to debate other drivers such as interest rates.
Team-shape options
- Same teams, AI-augmented: lowest disruption; gains depend on removing bottlenecks.
- Fewer, more senior engineers per team: higher review capacity and judgment; risks an empty junior pipeline and key-person dependency.
- Product teams + strong AI platform team: central team builds gateway, evals and agent pipelines; product teams consume them.
- Agent operators / orchestrators: engineers who run fleets of agents on maintenance and migration work.
The junior question
If you stop hiring juniors, where do your seniors come from in five years? Peng and colleagues' Copilot experiment suggested AI may especially help people newer to software development; and DX's Q1 2026 data shows junior engineers who use AI daily now saving about as much time each week as daily Staff+ users. A deliberate junior programme — paired with seniors, using AI to learn faster — can be an advantage.
Change management
Update career ladders, performance expectations and training. Measure before you reshape: two or more periods of scorecard data per team before cutting or merging.