Case study
Our client is a US-based automotive platform that facilitates used-car transactions by connecting sellers with an extensive buyer network, ensuring a smooth and efficient selling experience.
Understanding how users interact with the platform is critical for tracking performance, optimizing conversions, and improving the overall customer experience. To support this, the company invested in a general-purpose AI platform for autonomous data analysis.
Although the system produced accurate results, its execution method was inefficient for frequent use. Each request was handled as a single, end-to-end operation, regardless of complexity or size.
This led to:
As usage grew, both costs and latency increased, making the system unsustainable.
To address these challenges, the team at Maruti Techlabs redesigned the workflow, using n8n as the orchestration layer and adopting an agent-driven architecture.
Rather than processing each request as a single operation, the workflow was divided into smaller, specialized steps managed by dedicated agents:
Distributing the workload across these agents enabled incremental data processing, which improved efficiency, flexibility, and overall performance.
The redesigned workflow delivered immediate, measurable improvements in both cost and performance.
