Echelon

Arga Labs provides isolated sandbox environments equipped with API twins of third-party services, enabling developers and organizations to safely test and train AI agents without risking their production systems. By offering a controlled testing ground that mirrors real-world APIs, the company allows teams to experiment with AI implementations, validate workflows, and refine agent behavior in a risk-free setting. This approach addresses a critical need in AI development, where testing in live environments can introduce costly errors or security vulnerabilities.

Avallon creates AI agents that make claims operations teams an order of magnitude more efficient. The company is composed of engineers and operators with deep industry experience whose backgrounds span top institutions like Cornell, MIT, and HSG and whose work includes published research in NLP. Avallon has scaled technology in complex, traditional industries and built products trusted by Fortune 500 companies. The company grew revenue 10x in three months during Y Combinator and is backed by top investors including YC, Frontline Ventures, 1984, and Booom VC as it scales rapidly.

The company aims to make building affordable by leveraging AI to analyze blueprints. It focuses on building computer vision object detection models and coordinating these models, with an emphasis on ML and algorithm development. The company states it prioritizes the intelligence side (ML and algorithms) over the application side. Backed by investors who have funded PlanGrid, Dropbox, and Airbnb, it has strong backing. It is part of Y Combinator's Winter 2025 batch (YCW25).