It looks like nearly each week for the final two years since ChatGPT launched, new massive language fashions (LLMs) from rival labs or from OpenAI itself have been launched. Enterprises are arduous pressed to maintain up with the huge tempo of change, not to mention perceive tips on how to adapt to it — which of those new fashions ought to they undertake, if any, to energy their workflows and the customized AI brokers they're constructing to hold them out?
Assist has arrived: AI purposes observability startup Raindrop has launched Experiments, a brand new analytics function that the corporate describes as the primary A/B testing suite designed particularly for enterprise AI brokers — permitting firms to see and evaluate how updating brokers to new underlying fashions, or altering their directions and power entry, will influence their efficiency with actual finish customers.
The discharge extends Raindrop’s present observability instruments, giving builders and groups a strategy to see how their brokers behave and evolve in real-world situations.
With Experiments, groups can monitor how modifications — reminiscent of a brand new device, immediate, mannequin replace, or full pipeline refactor — have an effect on AI efficiency throughout hundreds of thousands of consumer interactions. The brand new function is accessible now for customers on Raindrop’s Professional subscription plan ($350 month-to-month) at raindrop.ai.
A Knowledge-Pushed Lens on Agent Improvement
Raindrop co-founder and chief expertise officer Ben Hylak famous in a product announcement video (above) that Experiments helps groups see “how actually something modified,” together with device utilization, consumer intents, and problem charges, and to discover variations by demographic components reminiscent of language. The purpose is to make mannequin iteration extra clear and measurable.
The Experiments interface presents outcomes visually, exhibiting when an experiment performs higher or worse than its baseline. Will increase in unfavorable alerts would possibly point out larger activity failure or partial code output, whereas enhancements in constructive alerts might mirror extra full responses or higher consumer experiences.
By making this knowledge simple to interpret, Raindrop encourages AI groups to strategy agent iteration with the identical rigor as fashionable software program deployment—monitoring outcomes, sharing insights, and addressing regressions earlier than they compound.
Background: From AI Observability to Experimentation
Raindrop’s launch of Experiments builds on the corporate’s basis as one of many first AI-native observability platforms, designed to assist enterprises monitor and perceive how their generative AI methods behave in manufacturing.
As VentureBeat reported earlier this 12 months, the corporate — initially often called Daybreak AI — emerged to handle what Hylak, a former Apple human interface designer, known as the “black field drawback” of AI efficiency, serving to groups catch failures “as they occur and clarify to enterprises what went improper and why."
On the time, Hylak described how “AI merchandise fail consistently—in methods each hilarious and terrifying,” noting that not like conventional software program, which throws clear exceptions, “AI merchandise fail silently.” Raindrop’s unique platform targeted on detecting these silent failures by analyzing alerts reminiscent of consumer suggestions, activity failures, refusals, and different conversational anomalies throughout hundreds of thousands of each day occasions.
The corporate’s co-founders— Hylak, Alexis Gauba, and Zubin Singh Koticha — constructed Raindrop after encountering firsthand the issue of debugging AI methods in manufacturing.
“We began by constructing AI merchandise, not infrastructure,” Hylak instructed VentureBeat. “However fairly rapidly, we noticed that to develop something severe, we would have liked tooling to know AI conduct—and that tooling didn’t exist.”
With Experiments, Raindrop extends that very same mission from detecting failures to measuring enhancements. The brand new device transforms observability knowledge into actionable comparisons, letting enterprises take a look at whether or not modifications to their fashions, prompts, or pipelines really make their AI brokers higher—or simply completely different.
Fixing the “Evals Move, Brokers Fail” Downside
Conventional analysis frameworks, whereas helpful for benchmarking, hardly ever seize the unpredictable conduct of AI brokers working in dynamic environments.
As Raindrop co-founder Alexis Gauba defined in her LinkedIn announcement, “Conventional evals don’t actually reply this query. They’re nice unit checks, however you possibly can’t predict your consumer’s actions and your agent is operating for hours, calling lots of of instruments.”
Gauba mentioned the corporate persistently heard a standard frustration from groups: “Evals cross, brokers fail.”
Experiments is supposed to shut that hole by exhibiting what really modifications when builders ship updates to their methods.
The device permits side-by-side comparisons of fashions, instruments, intents, or properties, surfacing measurable variations in conduct and efficiency.
Designed for Actual-World AI Habits
Within the announcement video, Raindrop described Experiments as a strategy to “evaluate something and measure how your agent’s conduct really modified in manufacturing throughout hundreds of thousands of actual interactions.”
The platform helps customers spot points reminiscent of activity failure spikes, forgetting, or new instruments that set off sudden errors.
It can be utilized in reverse — ranging from a identified drawback, reminiscent of an “agent caught in a loop,” and tracing again to which mannequin, device, or flag is driving it.
From there, builders can dive into detailed traces to seek out the foundation trigger and ship a repair rapidly.
Every experiment supplies a visible breakdown of metrics like device utilization frequency, error charges, dialog period, and response size.
Customers can click on on any comparability to entry the underlying occasion knowledge, giving them a transparent view of how agent conduct modified over time. Shared hyperlinks make it simple to collaborate with teammates or report findings.
Integration, Scalability, and Accuracy
In keeping with Hylak, Experiments integrates straight with “the function flag platforms firms know and love (like Statsig!)” and is designed to work seamlessly with present telemetry and analytics pipelines.
For firms with out these integrations, it might probably nonetheless evaluate efficiency over time—reminiscent of yesterday versus in the present day—with out extra setup.
Hylak mentioned groups usually want round 2,000 customers per day to provide statistically significant outcomes.
To make sure the accuracy of comparisons, Experiments screens for pattern measurement adequacy and alerts customers if a take a look at lacks sufficient knowledge to attract legitimate conclusions.
“We obsess over ensuring metrics like Process Failure and Person Frustration are metrics that you simply’d get up an on-call engineer for,” Hylak defined. He added that groups can drill into the precise conversations or occasions that drive these metrics, guaranteeing transparency behind each combination quantity.
Safety and Knowledge Safety
Raindrop operates as a cloud-hosted platform but in addition affords on-premise personally identifiable data (PII) redaction for enterprises that want extra management.
Hylak mentioned the corporate is SOC 2 compliant and has launched a PII Guard function that makes use of AI to mechanically take away delicate data from saved knowledge. “We take defending buyer knowledge very severely,” he emphasised.
Pricing and Plans
Experiments is a part of Raindrop’s Professional plan, which prices $350 per 30 days or $0.0007 per interplay. The Professional tier additionally consists of deep analysis instruments, matter clustering, customized problem monitoring, and semantic search capabilities.
Raindrop’s Starter plan — $65 per 30 days or $0.001 per interplay — affords core analytics together with problem detection, consumer suggestions alerts, Slack alerts, and consumer monitoring. Each plans include a 14-day free trial.
Bigger organizations can go for an Enterprise plan with customized pricing and superior options like SSO login, customized alerts, integrations, edge-PII redaction, and precedence help.
Steady Enchancment for AI Programs
With Experiments, Raindrop positions itself on the intersection of AI analytics and software program observability. Its give attention to “measure fact,” as acknowledged within the product video, displays a broader push inside the business towards accountability and transparency in AI operations.
Slightly than relying solely on offline benchmarks, Raindrop’s strategy emphasizes actual consumer knowledge and contextual understanding. The corporate hopes this can permit AI builders to maneuver quicker, determine root causes sooner, and ship better-performing fashions with confidence.
[/gpt3]
