Okahu x Paygentic: Project Monocle

Teams building AI and agentic applications need to solve two problems:

• Instrument their application to enable tracing and observability of their AI workflows.
• Export and process usage metrics to monetize their application using some pricing model that reflects consumption.

Traditionally this involved large amounts of custom code in the application’s hot path. Solutions and specs like OpenTelemetry and related technologies in the cloud native ecosystem standardized this for traditional applications, but GenAI introduced new complexities and requirements that could not be handled properly with existing exporters and frameworks.

Enter Monocle. Monocle helps developers and platform engineers building or managing GenAI apps to monitor them by making it easy to instrument their code and capture traces that are compliant with the open-source cloud-native observability ecosystem.

Monocle provides an implementation + package, not just a spec

• Works out of the box — no OpenTelemetry expertise needed to get started
• Ready-made implementation included — skip the work of building against the spec yourself
• GenAI-aware instrumentation built in — the domain-specific last mile is already handled

The project is actively maintained by the Okahu team and a growing community of contributors. Okahu is a platform built specifically for observing AI workloads with generative capabilities that use components such as OpenAI, Langchain, Nvidia Triton, Microsoft Azure and more. This means simpler onboarding, more automated discovery, and more relevant insights for improving your AI workload.

With Monocle, robust and performant telemetry can be achieved with a single line of code. And now the same line of code can turn that telemetry data into a monetization workflow thanks to the Paygentic exporter.

Paygentic offers a modular monetization stack designed for AI and agentic software applications. It includes metering, billing, and payment capabilities that scale together with your workloads, integrated into a unified full-stack solution.

Usage and consumption based pricing is becoming predominant in AI solutions. A question that we always get asked when starting an integration is how to properly instrument a system to export usage metrics to Paygentic. Now with Monocle and our custom Paygentic Exporter, we are happy to offer a standardized plug-and-play solution compatible with most existing GenAI frameworks such as Langchain and CrewAI. The same line of code that gives you immediate observability gains with Okahu can be used to track usage and charge for it through Paygentic.

‍

‍

Working with the Okahu and Monocle team was amazing for bringing this solution to the ecosystem. We hope that Monocle can make state-of-the-art observability and monetization available to any team building in the space, with minimal effort and maintenance.

If you want to learn more, take a look at the Monocle project and the Paygentic exporter. We are curious about what you will build with it.

‍

‍