AI agents bypass traditional user onboarding, immediately push software to its limits, and even request missing API functionality. This requires a fundamental shift in product development, focusing on robust, agent-friendly infrastructure rather than just human-centric UIs.
Klaviyo's internal "Dark Factory" system uses a team of agents to take a high-level prompt, decompose it into specs and engineering subsystems, write code, and establish API contracts. This automates the software toolchain, from prototyping to load testing, with minimal human intervention.
View a general-purpose LLM as a highly athletic but untrained high schooler. To make it great at a specific task, you must build a "harness" around it, providing specialized coaching, real-time data feeds, and feedback loops to develop its domain-specific expertise.
Klaviyo established a proficiency scale for AI, mandating that all product and design staff reach "Level 3." This means they are not just using AI for simple tasks but are constantly running their own agents or teams of agents to complete their work, ensuring deep, practical adoption.
To manage the risk of AI-driven feature bloat, Klaviyo compiled years of product review feedback into a database. An AI agent now consults this "taste database" to pre-vet new ideas, ensuring they align with company principles before they reach human review, scaling quality control.
AI agents make software UIs secondary. The real product is now the headless infrastructure layer of data and APIs that agents interact with. This means companies must shift their primary focus from pixel-perfect interfaces to building robust, accessible, and well-documented infrastructure.
To avoid the impossible task of teaching thousands of SMB customers to build AI agents, Klaviyo uses internal agents to automatically train and configure customer-facing agents. This "agent-trains-agent" model bypasses manual implementation, making sophisticated AI accessible at scale.
AI agents interact with software exclusively through APIs, bypassing the UI. This means a product with a dated front-end can become highly valuable and competitive in the agentic era if it exposes a robust, modern, and well-documented API layer that agents can leverage.