Monte Carlo Data launches Bayes, an AI agent for autonomously monitoring and fixing autonomous agents
Enthusiasm for artificial intelligence agents is at an all-time high and many enterprises have gotten so carried away that they’re deploying them faster than they keep up.
Their human AI engineers only have the capacity to watch over a handful of AI agents at any one time, so they’re in desperate need of help. What better solution could there be than more AI agents?
That’s what the big data observability company Monte Carlo Data Inc. seems to think anyway. Today it announced a new offering called Bayes, described as an “AI teammate” to differentiate it from the autonomous agents it’s designed to watch. Bayes is really an agent itself, but rather than do work itself, it’s tasked with observing the ones that do get work done.
It watches for signs of unreliability and if it spots something, it troubleshoots the problem, whether that’s with the agent itself or the data being fed into it, implements a fix autonomously and then verifies that the agent is now functioning as it’s supposed to.
Bayes is the culmination of Monte Carlo’s evolution from big data observability to AI. The company borrowed the observability concept from the world of software applications, pioneering tools to help humans keep tabs on their structured data environments. It later expanded to cover unstructured data too, before pivoting last year to embrace generative AI, launching tools for monitoring both inputs and outputs.
The company says there’s an urgent need for Bayes. The average human engineer struggles to monitor more than two or three AI agents and their associated data pipelines at a time, but many organizations have deployed many more than this. With each new agentic deployment, there’s a need to evaluate its work constantly and ensure that its outputs don’t start going astray later on. Clearly, it’s time to automate that task too.
Monte Carlo’s pitch has resonated with customers such as Axios Media Inc., the technology news website that uses AI agents for research, drafting, localization, managing news feeds and other tasks. Shreye Saxena, senior data scientist at Axios, explained that the nature of media work means it cannot compromise on its agent’s accuracy. “If the injected data is incorrect or the trajectory is wrong and it produces a response without checking the database, then we’re not going to get the outcomes we need from our agents,” he said.
Bayes works by scanning a company’s technology environment to discover the agents they’ve deployed and the data environments that power them. It can be given as much or as little responsibility as the customer feels comfortable with, performing tasks like monitoring agent’s outputs, triaging issues and executing fixes in the background.
When it spots a problem, it will trace that failure all the way back to the developmental decisions made pre-deployment, studying both the model that powers it and the data pipelines it relies on. To do that, it integrates with third-party coding assistants and software deployment pipelines.
Gartner Inc. analyst Sanjeev Mohan said Monte Carlo is effectively going after the Holy Grail of AI, giving enterprises a way to build self-improving AI agents. “The idea isn’t new, but pulling it off has been elusive,” he said. “You have to see what agents actually do at runtime and trace it back through the system all the way to the data.”
Monte Carlo says Bayes has already had a profound impact on early adopters. A South American financial services institution reported that Bayes detected 200% more incidents than its internal monitoring systems, while also reducing the number of false positives by 80%. The cloud data warehouse giant Snowflake Inc. is using Bayes to help its own customers build and run more reliable Cortex-based agents.
The goal, according to Monte Carlo co-founder and Chief Executive Barr Moses, is to make customers feel as if they can trust their AI agents. “There aren’t enough engineers to keep those agents trustworthy, but Bayes changes the math,” he said. “It understands your agents and your data together, fixes what breaks and gets better at your environment every day, so teams can spend less time firefighting and more time shipping.”
Image: SiliconANGLE/Meta AI
Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.
- 15M+ viewers of theCUBE videos , powering conversations across AI, cloud, cybersecurity and more
- 11.4k+ theCUBE alumni — Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network
Are you an AWS customer? Support SiliconANGLE financially by buying your AWS services from our Marketplace portal page and links: https://siliconangle.com/aws-marketplace/
About SiliconANGLE Media
theCUBE AIand theCUBE SuperStudios — with flagship locations in Silicon Valley and the New York Stock Exchange — SiliconANGLE Media operates at the intersection of media, technology and AI.
Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.