An entrepreneurship program by Canastra Ventures

São Paulo | Hybrid | June 1–26

The 4-week program designed to match exceptional Founders with top AI Talent to build AI-Native Services startups

In AI-native services, ideas are abundant. Breakthrough execution depends on exceptional AI-native teams.

Illustration: A firm in 2025, where humans still rely on outdated tools to handle repetitive tasks, while AI talent arrives to automate nearly everything. (Image generated with DALL·E 3)

Illustration: A firm in 2025, where humans still rely on outdated tools to handle repetitive tasks, while AI talent arrives to automate nearly everything. (Image generated with DALL·E 3)

Join a network of elite AI talent and founders

<aside>

200+

founders & AI talents

</aside>

<aside>

40+

founders matched

</aside>

<aside>

30+

startups created

</aside>

<aside>

40%

continued post-program

</aside>


The next $1T company will be a software company masquerading as a services firm.

Julien Bek — Partner, Sequoia Capital


ABOUT THIS EDITION

A new category is emerging: AI-Native Services

AI is not just optimizing services, it is redefining how services are designed, delivered, and scaled across industries. The next generation of startups will no longer look like SaaS companies; they will be AI-native services built around autonomous execution, continuous learning, and end-to-end value delivery.

<aside>

image.png

Legal Services

AI-native systems automating end-to-end legal workflows with human-grade precision.

</aside>

<aside>

image.png

Healthcare Admin

AI-driven operations layer removing administrative burden across clinical systems.

</aside>

<aside>

image.png

Marketing Execution

Autonomous execution engines turning strategy into continuous, compounding growth.

</aside>

<aside>

image.png

Enterprise IT

Self-operating infrastructure managing systems with minimal human intervention.

</aside>

The next wave of technology companies will no longer be defined by software as a tool, but by software as execution. We are entering a phase where software does not just enable work, but directly performs it end-to-end. This shift is taking place in a context where the global outsourcing market already exceeds $1T+, growing at ~6–8% annually, while AI services are emerging as a new multi-hundred-billion-dollar opportunity over the next 5 years. At the same time, 70–80% of companies are already using AI in at least one business function, signaling that adoption has moved from experimentation to structural transformation.

This transition unlocks the reconfiguration of trillion-dollar, labor-intensive markets such as finance operations, legal, IT, and customer support, where AI is already demonstrating the ability to reduce costs by up to 50% in specific functions and automate more than 50% of support tickets in certain use cases. For founders, this is not about building better software tools, but about redesigning entire categories under an “outcomes-as-a-service” paradigm, where value shifts from providing access to tools to delivering the final outcome directly. This shift, however, only becomes possible when deep domain operators in these services converge with a new generation of AI-native talent.


THE PROGRAM

Experience

that drives outcomes.

<aside>

Illustration: It’s 2026, and the new wave is building AI-native services startups — a $10+ trillion market. (Image generated with DALL·E 3)

Illustration: It’s 2026, and the new wave is building AI-native services startups — a $10+ trillion market. (Image generated with DALL·E 3)

</aside>

This Matchmaking edition is designed to connect deep domain operators in these services converge with a new generation of AI-native talent: those with a market vision and those who know how to build. It’s for people ready to go all-in on an AI-native services startup. In four weeks, each team walks away with a clear idea, a functional prototype, and enough clarity to decide whether to keep building together.

<aside>

</aside>


<aside>

</aside>


<aside>

</aside>


<aside>

</aside>


<aside>

</aside>