Archive · 24 Sep 2026
AI Reshapes Engineering Teams Into a New Hourglass Structure
The article says AI is changing how engineering teams are organized.
It does not simply mean that AI will replace all engineers.
Instead, some tasks that used to require several people can now be done faster with AI tools.
This may make the middle layer of coordinators and reviewers smaller.
Experts with deep knowledge remain especially important for difficult or regulated work.
Newer engineers can also accomplish more when they work with AI.
However, entry-level workers may have fewer opportunities to learn through routine tasks.
Leaders should plan for these changes and explain them honestly to their teams.
AI is compressing the traditional middle layer of engineering teams rather than evenly reducing every role.
The article describes a new hourglass structure with domain experts at the top and AI-augmented execution at the bottom.
AI makes translation, coordination, prototyping, testing, and other routine work faster and less labor-intensive.
Domain expertise remains highly valuable in physical and regulated fields where judgment and long-term context are essential.
Leaders are advised to plan workforce shape, reskill remaining coordinators, and communicate changes honestly.
- Who
- Engineering teams, operating leaders, domain experts, and AI-augmented engineers; the article was written by Mritunjay Singh, chief operating officer of L&T Technology Services.
- What
- AI is changing the structure and responsibilities of engineering teams into an hourglass-shaped model.
- When
- The article describes the change as already occurring, but gives no specific date.
- Where
- Engineering organizations and teams; no specific location is identified.
- Why
- AI has made much of the translation, coordination, and routine execution between ideas and finished work faster and cheaper.
- Central idea
- AI is changing the shape of engineering teams rather than uniformly replacing engineers.
- Team structure
- The proposed model is an hourglass with deep experts at the top, AI-augmented execution at the bottom, and a narrower middle.
- Middle-layer change
- Coordination, translation, reviewing, and handoff work are being compressed as AI reduces the distance between ideas and working prototypes.
- Expertise
- Domain judgment remains particularly important in physical and regulated fields.
- Entry-level impact
- Routine apprenticeship tasks such as first drafts, test cases, and boilerplate are increasingly automated.
- Leadership advice
- Leaders should plan workforce shape, strengthen higher-order coordination skills, and communicate changes candidly.
- Author
- Mritunjay Singh, chief operating officer of L&T Technology Services