It's Time to Redesign How Product Teams Work

I've been watching product teams work for a while now, and there's a pattern that plays out almost everywhere. A product manager writes a ticket. A designer creates mockups. An engineer builds the feature. QA finds the bugs. Ship it. Repeat.

It's like a relay race where the baton gets a little lighter at every handoff, and also a little blurrier. This workflow has defined software teams for decades. And honestly? It's falling apart faster than most people realize.

When AI Builds 90% of the Feature

Alexander Kirillov recently documented an experiment that should make every product leader sit up straight. Using autonomous coding agents, he found that AI could take him to 90% of a finished feature, with human refinement covering the final stretch.

Let that sink in for a second. Ninety percent.

That number isn't a fun curiosity. It's a signal flare. If most of the building can be handled by AI, then your product team's value doesn't live in the execution anymore. It lives in the orchestration. It's the difference between playing every instrument in the band and being the conductor.

The New Role Paradigm

Here's the thing about the traditional handoff chain (PM to designer to engineer to QA): it was already kind of leaky. Ideas got diluted at every step, context got lost in translation, and intent degraded through layers of interpretation. We've all played telephone and seen how that ends.

AI-driven workflows don't just speed up this chain. They compress it entirely. And that changes what every role actually looks like:

PM becomes Intent Architect. Writing Jira tickets becomes the least important part of the job. The real skill? Crafting precise context, defining guardrails, and articulating intent so clearly that an AI system, or a team orchestrating one, can execute with minimal drift. Think of it like writing really good instructions for someone who's incredibly capable but takes everything literally.

Engineer becomes Architectural Guardian. When AI handles the 80% that used to be boilerplate, engineers shift toward what only humans can do well: system design, security architecture, edge-case reasoning, and keeping an eye on AI-generated code that looks correct but might be subtly off. You know that code review where everything seems fine but something feels wrong? That instinct becomes gold.

Designer becomes Real-Time Reviewer. Static handoffs and pixel-perfect specs give way to live evaluation. Designers review AI-generated outputs as they emerge, adjusting on the fly rather than designing in isolation and hoping the implementation matches. It's more like sculpting than drafting blueprints.

QA becomes Quality Architect. Manual test cases take a back seat to designing automated, self-healing feedback loops. Quality assurance evolves from catching bugs after the fact to building systems that prevent them by design. It's a much cooler job, honestly.

This Doesn't Eliminate Roles. It Elevates Them

Here's the critical thing: none of these roles disappear. They move away from manual "doing" toward designing and orchestrating systems. The engineer who spends 80% of their day writing CRUD endpoints is underutilized. The same engineer designing the architecture that an AI implements? That's leverage. That's the fun part of the job.

But (and this is a big but) this shift requires something most organizations aren't ready for: compressed communication. When AI sits between intent and execution, the cost of being vague explodes. A fuzzy ticket that a senior engineer could interpret becomes a hallucinated feature when fed to an AI agent. Precision in how teams communicate (what they want, why they want it, and what "done" looks like) becomes a real competitive advantage.

I've seen this firsthand. The teams that write clear, specific briefs get amazing results from AI. The teams that write "make it better" get... well, you can imagine.

Why This Matters Now

The companies that figure this out first will ship faster with smaller teams. The ones that don't will keep adding headcount to solve problems that are fundamentally about workflow design, not capacity. It's like trying to fix a traffic jam by buying more cars.

At Heimdall, we work with businesses navigating exactly this transition. The technology is ready. The question is whether your team structure and workflows are ready to meet it.

The shift from execution to orchestration isn't coming. It's here. The only choice is whether you redesign your workflows deliberately, or let the disruption redesign them for you. And trust me, you'd rather be in the driver's seat for that one.

---

Want to explore how AI-driven workflows could reshape your product team? Get in touch and let's map out what the transition looks like for your organization.