The New Intrapreneurs

Hello 👋

It’s Thursday, 12^th November 2026. Welcome to Bold Efforts!

A few years ago, if a large company wanted to become more entrepreneurial, it might hire an Entrepreneur in Residence (EIR).

The idea was to bring a founder's mindset into an established organization. Give someone the freedom to spot opportunities, challenge assumptions, experiment, and perhaps build an entirely new business using the company's resources.

These roles still exist. But there's an interesting new variation emerging, driven by artificial intelligence.

Instead of hiring entrepreneurs to build new businesses, some companies are assembling small teams to reinvent the businesses they already have. And rather than working like traditional corporate transformation teams, these groups increasingly resemble the forward-deployed engineering teams found at companies like Palantir and OpenAI.

I think this could become one of the more interesting organizational changes of the AI era.

If you've been following enterprise AI, you've probably encountered the term forward-deployed engineer.

The concept isn't new. Palantir pioneered a model in which engineers work closely with customers, understand their problems, and build solutions inside the messy reality of their organizations.

AI has made this approach particularly relevant. Companies can buy access to powerful models relatively easily. Getting those models to work reliably inside actual business processes is a different challenge.

That's why companies like OpenAI have expanded their deployment capabilities, while IBM has introduced Forward Deployed Units, small multidisciplinary teams that work directly with clients.

These teams don't simply recommend what a company should do. They help build and implement it.

But what happens when the customer decides to build its own forward-deployed team?

Instead of repeatedly bringing in external specialists, an enterprise hires its own small group of strategists, engineers, and operators. They move between departments, understand problems, build AI-powered solutions, and help teams adopt new ways of working.

They're effectively forward-deployed engineers and strategists, except their customer is their own employer.

Imagine a global hotel company operating 150 properties.

Its finance department spends an extraordinary amount of time collecting data from individual hotels, reconciling inconsistent reports, and preparing monthly performance reviews. It's an old problem. Everyone knows about it. Several improvement projects have been proposed over the years.

Traditionally, the company might commission a consulting study, define requirements, obtain budget approval, and begin a technology implementation.

Now imagine assigning a three-person internal team to the problem.

One strategist who understands business operations and can work with the finance leadership. Two engineers who can build AI-powered workflows, connect systems, and turn ideas into working software.

They spend time with the people doing the work. They discover which steps are genuinely necessary, which exist because of outdated systems, and where AI could help. They build something small, test it with actual employees, and keep improving it.

Importantly, their job doesn't end when the software works. It ends when the business process works better.

Once the solution is running reliably and the finance team can own it, the pod moves on. Perhaps to procurement, revenue management, or guest services.

Over time, a company could have several such teams working across its business, sharing common tools and lessons.

The three-person configuration is illustrative, of course. Some problems need more specialists, and not every project can be completed in a few weeks. What matters is the model: small teams with enough business judgment, technical capability, and authority to take responsibility for outcomes.

Small cross-functional teams aren't a new invention. Neither are internal innovation labs or corporate transformation offices.

What's changing is the amount of work a small team can realistically accomplish.

AI-assisted software development makes it faster to prototype applications, write integrations, test ideas, and automate repetitive tasks. A capable engineer can now accomplish things that previously required considerably more time or a larger development team.

At the same time, many enterprise AI problems aren't primarily about inventing new technology. They're about applying existing capabilities to highly specific workflows.

Consider what this means for a company with thousands of employees and hundreds of processes. It may not need a giant team to reinvent everything simultaneously. It could build a small internal capability that tackles important problems sequentially, learning and improving with each deployment.

We're already seeing early signs of this shift.

Travelers, the American insurance company, has described an operating model where central AI specialists work with cross-functional business teams, embedding engineers to solve problems and transfer expertise. Meanwhile, companies including Target, Barclays, and American Express have advertised for forward-deployed engineering talent directly.

External deployment teams aren't disappearing. But the capability is beginning to move inside the enterprise.

This is where the comparison with Entrepreneurs in Residence becomes particularly interesting and sensible.

Traditional corporate EIR roles often focused on exploring new markets, developing products, or creating businesses adjacent to the parent company.

These new internal deployment roles have a different mission. Their entrepreneurial opportunity already exists inside the organization.

They don't need to find customers, raise capital, or invent a market. They inherit something many startups would love to have: real customers, valuable problems, existing infrastructure, and access to substantial resources.

Of course, they also inherit bureaucracy, legacy technology, internal politics, and employees who may not particularly want their jobs redesigned.

And that's precisely why the strategist in this model matters as much as the engineers.

Someone needs to understand which problems are worth solving, persuade business leaders to participate, navigate competing priorities, and decide what success actually means. Building the technology is only part of the job.

The most effective people in these roles may combine the curiosity of a founder, the commercial judgment of a strategist, the practicality of an operator, and enough technical understanding to work closely with engineers.

They might spend one quarter improving financial operations, another redesigning customer support, and the next working on supply chains.

It's a career that looks less like climbing a conventional corporate ladder and more like building a series of ventures within the same organization.

There's an obvious risk here.

A company could create ten exciting AI teams, give them impressive titles, and end up with ten disconnected prototypes that nobody maintains.

Small teams are only effective when they have real ownership, access to systems and data, support from business leaders, and a clear path for transferring their solutions into everyday operations.

They also need a common technical foundation. Otherwise, every pod will rebuild the same integrations, authentication systems, and monitoring tools.

And crucially, not every business problem needs a custom AI solution. Sometimes the right answer is better software, a simpler process, or eliminating unnecessary work altogether.

The objective shouldn't be to maximize the number of AI projects. It should be to improve how the company operates.

That requires giving these teams something traditional innovation initiatives often lack: accountability beyond the demonstration.

For a long time, the popular story was that ambitious, entrepreneurial people would leave large organizations to build something new. Companies responded by creating venture studios, incubators, and EIR programs to capture some of that energy internally.

AI creates another possibility.

An entrepreneurial person may no longer need to start a new company to build meaningful technology, solve difficult problems, and see the impact of their work. They could do it repeatedly inside an organization that already operates at enormous scale.

And for enterprises, this could offer a different way to approach transformation. Instead of treating innovation as an occasional initiative, they could develop a permanent capability to reinvent parts of themselves.

The future of corporate AI transformation may not look like one enormous program with hundreds of people.

It might look like a handful of unusually capable teams, moving through the organization one problem at a time.

Not quite startups. Not quite consultants. Not quite traditional employees. Something closer to entrepreneurs who already have a company. Thank you for reading.

Best,
Kartik

I write Bold Efforts every week to think clearly about where work and life are actually headed. If you want these essays in your inbox, you can subscribe here.