LinkedIn just killed their APM (Associate Product Manager) program and replaced it with “Product Builder” training—teaching product management, design, AND engineering all at once. Not a pivot. A wake-up call.
The traditional division of labor is collapsing. For decades, we’ve operated with clean boundaries: Product Managers define the what, engineers build the how, designers own the look and feel. Specialization. Efficiency. Clear swim lanes.
But in 2026, I’m watching something different emerge at OpenAI, Anthropic, and a wave of AI-first companies: Product Engineers who own the full loop from idea → code → customer. Not “full-stack developers” (which is about technical breadth), but full-stack product ownership—owning outcomes, not just outputs.
The Numbers Don’t Lie
Open product engineering roles jumped 53.6% in 2026. OpenAI, Anthropic, and Meta are hiring “AI Product Engineers” at $500K-$700K—and they still can’t fill the roles fast enough. These positions sit at the ML-product boundary, requiring:
- Ownership of probabilistic systems (not deterministic code)
- Evaluation engineering as a first-class discipline (not just “ship and pray”)
- AI-assisted execution (using AI to build AI products)
Forward Deployed Engineers at OpenAI embed with Fortune 500s to build custom, production-grade solutions—they’re not just writing code, they’re owning the customer relationship and the business outcome.
The question isn’t “can you code?” or “do you understand product strategy?” It’s “can you own the entire customer problem end-to-end?”
What This Means for Traditional PMs
I’m a VP of Product at a Series B startup. Non-technical background, came up through consulting → APM at Google → Senior PM at Airbnb → now leading product strategy. I’m good at what I do: customer development, go-to-market, competitive positioning, cross-functional alignment.
But here’s the uncomfortable question I’m sitting with: If engineers can now own product decisions (with AI tools removing technical friction), what’s my value prop?
I used to translate between customer problems and technical solutions. I used to be the bridge. But if Product Engineers can see the customer problem AND build the solution AND measure the outcome… am I the bridge, or am I the bottleneck?
The Fork in the Road
I see two possible futures:
Path 1: Specialized Product Roles — AI PM, Growth PM, Platform PM. PMs get more technical and domain-specific, owning narrow but deep problem spaces.
Path 2: Generalized Product Ownership — Product Engineers own outcomes, PMs disappear or become “Product Builders” (LinkedIn’s bet).
At my startup, we’re still hiring traditional PMs and traditional engineers. But our best hires? The ones who cross boundaries. Engineers who ask “should we build this?” not just “how do we build this?” PMs who can read code reviews and challenge technical assumptions.
I’m wondering if we’re optimizing for a 2019 org structure when the market is screaming that 2026 looks different.
Questions for the Forum
I’d love to hear from this community:
-
Has anyone made this transition—from PM to Product Engineer, or from Engineer to Product ownership? What did you gain? What did you lose?
-
Are your engineering orgs hiring for this hybrid role, or are you keeping PM/Eng boundaries separate? What’s driving that decision?
-
What breaks at scale if everyone owns “everything”? Do we lose the benefits of specialization? Does this only work for 10-person teams?
-
Is this AI-company specific, or is this coming to all product organizations eventually?
I’m not convinced the PM-Engineer split is dead. But I’m also not convinced it’s the future. Help me think through this.
Sources: