Sierra's AI Strategy in a Nutshell

Thoughts on Sierra AI and risk factors for application layer AI startups

A common question in AI circles: where will the value actually accrue—infra or apps? And if it’s apps, which verticals will matter the most?

To answer this question, one startup I monitor closely is Sierra.

Many (mis)categorize Sierra as a “customer support AI” startup, but they are missing the point completely.

To borrow Brett Taylor (CEO of Sierra)’s words, what Sierra sells is an AI agent platform for building “customer-facing” AI agents. This means Sierra’s vision is explicitly to land-and-expand to handling all customer interactions for a given company. Basically, customer support is just the wedge.

But Sierra’s ambitions make enterprise adoption complicated. Essentially, the “ask” is for enterprises to offload its CX to an external agent—raising inevitable concerns around change management and vendor lock-in. I’ll unpack those later.

So what makes Sierra worth monitoring? That’s because it represents the best-case-scenario comp for AI agent startups that sell to enterprise. If Sierra can’t make it, it doesn’t bode well for other vertical AI startups which are basically run some variation of Sierra’s strategy.

Why is Sierra special? On paper, Sierra’s got everything going for them. It checks all the boxes: elite founders, top-tier investors, and a deep war chest. It sells work, not infra.

It also follows the AI agent playbook to the “tee”: outcome-based pricing, a hybrid low-code/SDK platform, and an enterprise GTM motion powered by forward deployed engineers (FDEs). Basically every “vertical AI automation” company does this.

In other words, Sierra is a live experiment in enterprise AI company-building: can enterprise AI agent companies succeed as standalone businesses, or are they destined to become acqui-hired like Adept, Character.AI, or Windsurf?

In this post, I’ll cover:

• a high level overview of Sierra’s business, and my thoughts on its strategy

• why Sierra’s outcome could shape how enterprise AI agent startups are perceived

• the key risk factors that might still derail the company

What business is Sierra in?

Sierra was launched in early 2024 by Brett Taylor (former Salesforce co-CEO and OpenAI board chair) and Clay Bavor (former VP of AR/VR at Google).

On the surface, it looks like just another enterprise chatbot or agent-assist platform company. Its early traction—logos like ADT, Sonos, CLEAR, etc—shows it’s resonating with consumer businesses that experience cyclical support volume and high seasonality.

Note, even at AWS (where I worked on the tech side of GTM), selling “agent-assist” or “contact-center software” was all about winning high volume / high seasonality customers.

So Sierra’s customer mix isn’t anything special or surprising. In any case, Sierra reportedly hit $20M ARR within its first year, and if current growth holds, it’s likely north of $50M by mid-2025.

But exactly what business is Sierra in?