How should growth companies structure AI work?

We recommend that growth companies build AI into how the business operates rather than layering it on top, through several organizational moves. First, give the work a home. Connect the systems that hold company knowledge, define repeatable workflows tied to customers and revenue, and put the AI capability where people already work. Second, put someone in charge. A model we see working well pairs a small central group, accountable for reliability and quality, with champions inside each function; the combined team is funded with real budget, and progress is measured against the operating metrics each workflow is meant to move. Adoption on its own is a cost, not a result. Finally, treat data readiness as the price of admission, fixing the data each priority project needs rather than waiting for everything to be perfect.

“The scarce skill is no longer knowing how to use these tools; the tools are increasingly table stakes. It is knowing how to change the way your company structures its work around them,” says Ashwin Subramani, Summit’s Head of AI. “That is an organizational capability, and it needs to be built as deliberately as any product.”

How is AI changing hiring and talent development models?

When companies reorganize work, roles change too. The companies we see managing this transition well are backfilling capabilities, not roles. Before replacing a departure, these companies ask which tasks have disappeared, whether the role still needs to exist and what capability to add instead. The logic applies to both experienced and entry-level roles. For early-career professionals, we’ve seen AI absorb routine work that has traditionally defined junior roles, but rather than eliminating these roles, it is changing what these companies are hiring for. Employers surveyed by the National Association of Colleges and Employers this spring reported that more than one-third of entry-level jobs at their respective companies now require AI skills — nearly triple the share reported in fall 2025.⁽³⁾ And because capabilities can be built as well as bought, many companies are investing inward; the companies BCG identifies as furthest along in AI maturity direct approximately 60% of their AI budgets to training and upskilling, versus roughly 25% at other companies.⁽⁴⁾

“You cannot hire your way to an AI-ready organization,” says Dayton Ogden, Head of Talent & Recruiting at Summit Partners. “The companies we see doing this well are explicit about the capabilities they will need next, and they are focused on developing many of those skills internally, even as they recruit.”

Hiring is being rebuilt on the same principle. “The interview has to look like the job,” says Tim Kohn, Summit’s Technologist-in-Residence. “Give candidates access to the tools they will use every day on the job, and watch how they frame the problem, verify the output and defend their decisions. That is where I’ve seen strong engineers distinguish themselves.”

What role should leadership play in AI adoption?

Leaders should be among the most active AI users in their companies. Those companies we see making meaningful progress have leaders who use the tools themselves, and that hands-on use can help them judge what is real and model the change they are asking for. It can also compound what good leaders already have — namely, deep functional knowledge — which better informs where AI should change the work. The rest of the job is structural, hinging on decisions only leadership can make: naming one accountable owner, defining what proficiency means role by role, and making successes visible enough that other teams copy them.

Is your company organized for AI?

Readiness shows up when AI moves from individual experimentation to shared, company-wide workflows. Summit-backed CoderPad offers a useful example. The company, whose platform hosts technical hiring for more than 4,000 businesses,⁽⁵⁾ began its AI effort the way many growth companies do: licenses and training for everyone. Adoption appeared healthy, yet fewer than 5% of employees built anything their colleagues could reuse; the gains made by motivated individuals stayed personal, at risk of walking out the door with the people who created them.⁽⁵⁾ CoderPad’s response was organizational rather than technical: a shared system with a named owner, connected to company knowledge, with workflows designed for specific jobs. Within four and a half months, 85% of employees were active users.⁽⁵⁾ “The AI change is not an IT rollout,” CEO Amanda Richardson told our Executive AI Forum this fall. “This is a culture change.”

Heading into 2027, we believe growth-stage management teams should consider the following organizational questions: Who owns AI in your company, and does that owner have real budget? What does AI proficiency mean for each role? And what role are you, as a leader, playing in AI adoption and value creation? The companies best prepared to answer may not be the ones that adopted AI first, but the ones that reorganized around it.

For more of our perspective on how growth companies are navigating AI — including key learnings from Summit's 2026 Executive AI Forum and a closer look at the technical talent market — visit our Resources page.