Do You Need a CDO, a CTO, or Neither — Yet?

You know your industry. You know the customer's problem better than anyone. You've raised a seed round to build a physical AI product — and now everyone is telling you to "hire a strong technical leader".

But which one? And do you need them full-time yet?

Three different jobs

These titles get used loosely. In a physical AI startup, they're genuinely different jobs:

Role	Owns	You need it most when
Head of Data & AI (CDO)	What data to collect, how it moves, which AI to build on it, data quality	Your product's value depends on data from devices you don't yet collect or use well
Head of Product (CPO)	What to build first, for whom, and what a pilot must prove	You have deep domain knowledge but no clear path from idea to a product customers pay for
Head of Engineering (CTO)	Architecture, technical hiring, delivery rhythm, reliability	You need to build and run a team that ships, and you can't judge engineers yourself

Many founders hire a CTO first by default. In physical AI, that's often the wrong order — because the hardest early decisions are about data and product, not about code.

Signs you need data and AI leadership first

• You're not sure your data is good enough for AI, or what you should be collecting
• Pilots stall because the model behaves differently on real customer equipment
• Investors ask about your data moat and you don't have a crisp answer
• Engineers keep proposing different stacks and you can't tell who's right

Signs you need product leadership first

• You have customer interest but no agreed definition of the first version
• The roadmap keeps growing and nothing ships
• You can't say what a successful pilot looks like in numbers

Signs you need engineering leadership first

• You already know exactly what to build, and the bottleneck is building it
• You need to hire several engineers and can't assess them
• Releases are slow, fragile or unpredictable

Why "full-time" is often premature

A senior leader who has done this at scale is expensive and takes months to hire — often longer than your runway can comfortably absorb. And at seed stage, the job is usually spiky: intense for a few weeks around key decisions, lighter in between.

That's the case for fractional leadership: senior judgment part-time, from week one, while you learn which full-time role you actually need. A good fractional leader should also help you hire your first full-time team — and make themselves less necessary over time.

And sometimes: neither, yet

If you haven't spoken to enough customers to know what to build, no technical leader will fix that. Spend the next few weeks on discovery. Come back when you can describe the first customer, the decision your product helps them make, and what a successful pilot looks like.

A simple way to decide

Write down the three decisions you're most worried about getting wrong in the next six months. If they're about data, start with a Head of Data & AI. If they're about what to build, start with product. If they're about building and shipping, start with engineering.

Not sure? That's what our Services page is designed to help with — or tell us what you're building and we'll give you an honest read, even if the answer is "not us".