Agentic Commerce Platform Launched
Yuno has launched a new commerce platform designed to help merchants monetize AI agent activity. The platform points toward a future where agent-driven transactions and recommendations become direct revenue channels. This development suggests growing demand for APIs and analytics that can interface with agentic commerce systems.
- Yuno’s platform provides a suite of developer tools including SDKs for Web, Android, and iOS, a server-to-server direct API integration, and options for PCI-compliant secure fields. The company also offers Postman collections for testing and development of its APIs, which cover payments, subscriptions, and reporting. - The NOVA AI agent suite is designed to recover revenue from failed transactions by engaging customers through WhatsApp or voice calls in over 70 languages. Early pilots with global merchants showed that NOVA helped recover up to 75% of failed payments on answered calls, and delivery company Rappi saw an 8% lift in its recovery rate within months of adoption. - The rise of agentic commerce is leading to new open standards for developers, such as the Agentic Commerce Protocol (ACP) by Stripe and OpenAI, and the Universal Commerce Protocol (UCP) from Google. These protocols define standards for how AI agents and businesses transact, covering checkout processes and the secure sharing of payment credentials through REST APIs and delegated payment tokens. - The underlying AI technologies in agentic commerce share parallels with quantitative finance. Both fields utilize Natural Language Processing (NLP)—in commerce for conversational interfaces, and in trading for analyzing market sentiment from news and social media to inform algorithmic trading strategies. - Both payment orchestration and high-frequency trading (HFT) are critically dependent on low-latency, real-time data processing. While HFT uses this for microsecond-level trade execution, payment platforms like Yuno use it for smart routing, which directs transactions to the optimal processor based on real-time performance to increase approval rates. - Machine learning models are core to both modern payment fraud detection and quantitative analysis. In payments, algorithms like logistic regression, random forests, and neural networks are used to identify anomalous transactions in large datasets. In quantitative trading, similar AI techniques are applied to recognize complex patterns in market data to predict behavior and manage risk.
