Agentic development is changing the economics of enterprise software. As applications become faster and less expensive to build, organizations can rethink where custom software makes sense and where standardized platforms still provide value.

But building more software introduces a different challenge. Applications and agents still need identity, data, workflows, security, observability, deployment and governance underneath them.

At the same time, agents are changing how teams interact with that infrastructure. More of the complexity involved in development and operations can be handled by agents, tooling and development harnesses rather than individual developers.

Nick Van Weerdenburg and Ben Hofferber will explore how these shifts come together in a post-SaaS application architecture — from Kubernetes and infrastructure to harness engineering and the operating layer used to build, deploy and manage what sits on top.

From dependency to ownership#

When does vendor lock-in become a business problem?#

When a vendor stops enabling the business and starts deciding what it can change. Buying software is usually the right call. Lock-in becomes a problem when a platform sits inside what differentiates you, such as your core product, customer experience or key workflows, and every change means a workaround or waiting on someone else's roadmap.

When should you consider alternatives to Salesforce?#

When its cost, constraints or dependency start shaping business decisions, and you can't credibly answer "what would it take to leave?" Before any Salesforce migration or replacement, map what your business actually uses: the data and workflows it depends on, what would be genuinely hard to move, and what is simply accumulated platform. That tells you whether the realistic alternative is another platform, owning a few key capabilities, or a phased migration, and lets you reduce SaaS dependency one capability at a time instead of replacing everything at once.

When should a business build vs. buy software?#

Buy what is commodity; own what differentiates. Specialized capabilities where a vendor's expertise and scale carry real operational, reputational or technical risk are usually still better bought. Capabilities tied to your customer experience, economics or competitive edge are increasingly worth owning, because AI-assisted development has made building and running them far more realistic. The goal isn't to build everything. It's to own the software that defines your business.

What you'll learn#

• How agentic development is changing the economics of building enterprise software
• How to think about build vs. buy as the cost of software development changes
• What infrastructure agentic applications need underneath them to operate in production
• How infrastructure, development harnesses and an operating layer fit together
• What a post-SaaS enterprise application architecture could look like going into 2027

This session is for you if…#

• You're rethinking what your organization should build versus buy as agentic development changes the economics of software.
• You're building or moving agentic applications into production and thinking about the infrastructure required underneath them.
• You're responsible for technology, product, engineering or architecture and thinking about how your enterprise stack needs to evolve.

This session aired live on November 10, 2026, and the recording is in post-production. Here is what the hour covered, and where to pick up the thread while you wait for it.

From dependency to ownership#

When does vendor lock-in become a business problem?#

When a vendor stops enabling the business and starts deciding what it can change. Buying software is usually the right call. Lock-in becomes a problem when a platform sits inside what differentiates you, such as your core product, customer experience or key workflows, and every change means a workaround or waiting on someone else's roadmap.

When should you consider alternatives to Salesforce?#

When its cost, constraints or dependency start shaping business decisions, and you can't credibly answer "what would it take to leave?" Before any Salesforce migration or replacement, map what your business actually uses: the data and workflows it depends on, what would be genuinely hard to move, and what is simply accumulated platform. That tells you whether the realistic alternative is another platform, owning a few key capabilities, or a phased migration, and lets you reduce SaaS dependency one capability at a time instead of replacing everything at once.

When should a business build vs. buy software?#

Buy what is commodity; own what differentiates. Specialized capabilities where a vendor's expertise and scale carry real operational, reputational or technical risk are usually still better bought. Capabilities tied to your customer experience, economics or competitive edge are increasingly worth owning, because AI-assisted development has made building and running them far more realistic. The goal isn't to build everything. It's to own the software that defines your business.

What the session covered#

• Why the traditional tradeoff between infrastructure capability and convenience is changing.
• What infrastructure agentic applications increasingly need underneath them: identity, data, workflows, observability and security.
• Where Kubernetes fits into the emerging agentic application stack.
• How harness engineering can make infrastructure capabilities accessible to development agents.
• How the infrastructure, development and operating layers come together in a post-SaaS architecture.

Keep reading while you wait#

• The Fast Fashion Era of SaaS: How Agentic Development Disrupted Our Business Model, on how agentic development is changing the build-versus-buy calculation.
• Browse upcoming Rangle events for the next live session.

The full session on the post-SaaS application architecture: the infrastructure applications run on, and the development harnesses and operating layer used to build and manage them.

From dependency to ownership#

When does vendor lock-in become a business problem?#

When a vendor stops enabling the business and starts deciding what it can change. Buying software is usually the right call. Lock-in becomes a problem when a platform sits inside what differentiates you, such as your core product, customer experience or key workflows, and every change means a workaround or waiting on someone else's roadmap.

When should you consider alternatives to Salesforce?#

When its cost, constraints or dependency start shaping business decisions, and you can't credibly answer "what would it take to leave?" Before any Salesforce migration or replacement, map what your business actually uses: the data and workflows it depends on, what would be genuinely hard to move, and what is simply accumulated platform. That tells you whether the realistic alternative is another platform, owning a few key capabilities, or a phased migration, and lets you reduce SaaS dependency one capability at a time instead of replacing everything at once.

When should a business build vs. buy software?#

Buy what is commodity; own what differentiates. Specialized capabilities where a vendor's expertise and scale carry real operational, reputational or technical risk are usually still better bought. Capabilities tied to your customer experience, economics or competitive edge are increasingly worth owning, because AI-assisted development has made building and running them far more realistic. The goal isn't to build everything. It's to own the software that defines your business.

What's in the recording#

• Why the traditional tradeoff between infrastructure capability and convenience is changing.
• What infrastructure agentic applications increasingly need underneath them: identity, data, workflows, observability and security.
• Where Kubernetes fits into the emerging agentic application stack.
• How harness engineering can make infrastructure capabilities accessible to development agents.
• How the infrastructure, development and operating layers come together in a post-SaaS architecture.