Larridin Blog

The AI-Native Engineering Playbook: Building AI-Native Teams

Written by Larridin | Nov 16, 2026

In October, Larridin was a platinum sponsor of the Engineering Leadership Conference (ELC) at Fort Mason Center in San Francisco, and our CTO, Ameya Kanitkar, took the main stage with a keynote called The AI-Native Engineering Playbook. The question underneath it is one he's been working through in public for a while: how should an engineering org be organized for AI?

He explained his thinking on LinkedIn:

Read Ameya's post on LinkedIn

Ameya Kanitkar, co-founder and CTO of Larridin, on LinkedIn:

Engineering leaders are reorganizing teams for AI and making completely opposite changes.

In our latest report, 40.8% said they had reorganized teams around products or outcomes over the past year.

But there was no common pattern.

Some made teams smaller. Others made them bigger.

23.5% widened manager spans. Exactly 23.5% narrowed them.

Some companies created more tech lead roles. Others added engineering managers. Some removed engineering manager roles entirely.

19.4% changed the mix of senior and junior engineers. Another 15.3% introduced roles focused on verification, agent platforms, or AI enablement.

And 29.6% had not changed their organization at all.

The answers are different because companies are running into different problems.

If agents allow a small team to own more, smaller teams may work.

If agents increase code output but create more work in review, architecture, and coordination, companies may need more technical leadership.

If AI adoption is uneven, the immediate need may be enablement rather than a new team structure.

There is no standard engineering organization for the AI era yet.

Companies are changing one part of the organization, watching what happens, and adjusting again.

Why a Playbook, and Why Now

Engineering leaders are being asked to redesign their orgs for AI, and there's no standard approach. Our AI Impact in Engineering Report shows companies making opposite moves on team size, manager spans, and roles, all in the same year.

So a useful playbook can't hand you an org chart. What it can do is help you pick the right change for the problem in front of you.

The Data: No Common Pattern

Here's how engineering leaders in the September 2026 LDX3 New York poll said their orgs changed over the past year. Approximately 98 people responded, and multiple selections were allowed:

  • 40.8% reorganized teams around products or outcomes
  • 23.5% widened manager spans, and exactly 23.5% narrowed them
  • 19.4% changed their mix of senior and junior engineers
  • 15.3% introduced roles focused on verification, agent platforms or AI enablement
  • 29.6% hadn't changed their organization at all

Step 1: Name the Problem You Actually Have

Ameya's point is that the answers differ because the problems differ. Three of the most common:

  • Agents let a small team own more. Smaller teams may work.
  • Agents boost code output, but review, architecture, and coordination can't keep up. You may need more technical leadership.
  • AI adoption is uneven across the org. Enablement may matter more right now than a new structure.

Most orgs have some mix of all three, so it helps to figure out which one is biting hardest before touching the org chart.

Step 2: Change One Thing at a Time

Ameya's advice is to change one part of the organization, watch what happens, and adjust again. Treat each change like an experiment, with a clear reason for making it and a window to see whether it worked.

Step 3: Measure Before the Next Move

The experiment only works if you can see the results. A few signals worth watching, depending on which problem you started with:

  • Whether smaller teams are shipping more or simply stretched thinner
  • How long work waits in review, and whether senior engineers are buried in it
  • How evenly AI use and AI fluency are spread across engineers

If the numbers don't move, that's useful too. It tells you the next change should target a different problem.

Closing Thought

There's no standard engineering organization for the AI era yet, and that's okay. The playbook is a loop: name the problem, change one thing, measure, and adjust.

If you caught Ameya at ELC or stopped by the Larridin booth, thanks for the great conversations.

Ameya shares this kind of practical, data-first stuff on LinkedIn pretty regularly. If you want more of it in your feed, follow Ameya Kanitkar on LinkedIn. He's co-founder and CTO of Larridin.

On larridin.com: