AI engineers use Raindrop to get alerts about hidden issues and successes in their AI products. Raindrop sends alerts when your AI misbehaves and links straight to events, so you can dig into conversations or traces, understand the root cause, and fix it, fast
Hey PH! We’re excited to launch Raindrop. You can sign up for Raindrop to get issue alerts for your AI product.
AI products fail constantly—in ways both hilarious and terrifying.
Regular software throws exceptions. But AI products fail silently.
Raindrop is the first Sentry-like monitoring platform for AI products.
The Problem Traditionally, when a user hits an error, it’s easy to detect and easy to get notified once it becomes an issue (this is exactly what Sentry does!). But when building AI products, these issues go undetected.
The current status quo is sifting through millions of logs and trying debug flaky evals.
Evals are not enough - like unit tests, they confirm that your model passed specific test cases. But in the real world, AI chatbots and agents encounter millions of unpredictable actions. AI engineers need issue monitoring to discover production issues so they can make AI products that not only pass tests but also perform well in the real world.
Solution Raindrop sends you alerts when your AI misbehaves and links straight to the events, so you can dig into the conversations or traces, understand the root cause, and fix it, fast.
Daily Alerts Include - Issues: detects issues like Assistant Forgetting, Laziness, Task Failure, User Frustration, and more, depending on the type of AI app - Wins: surfaces what your product does well, so you can double down on those experiences and create great evals
The Pro tier lets you go even further with:
- Custom Issues / Topics: define and track any issue or topic - Topic Clustering: clusters data in real-time to find your AI product’s most popular use cases, and what use cases have the most issues - Signals: finds patterns in explicit signals like thumbs up / thumbs down - Deep Research: deep research for your production AI data, letting you use natural language to search for any kind of behavior - Traces: track every step of your AI call - Edge PII Redaction: intelligently strip PII from any user and model messages - Dataset Creation: create custom datasets out of any set of events
Companies like Clay.com, Tolans, Websim, and more have been using Raindrop to improve their AI products. They’ve been able to quickly iterate on fixes, see how issue incidence rates decrease in production, and confidently ship changes.
“Raindrop has been invaluable as we've been growing quickly. It's critical for us to keep issue incidence below an acceptable threshold and become aware of any spikes. It's like if we see an iOS crash report in Sentry, but for our AI capabilities.” -Evan Goldschmidt, CTO Tolans
It’s like having a Sentry tailored for AI, which is super needed right now! How does it differentiate itself from traditional observability tools like Datadog or Sentry when it comes to AI-specific issues?
As someone who has worked on one of the largest deployed LLM products, our team would have loved to have something like this.
Observability has been a major challenge in every AI application I've worked on. Failure modes are varied and unpredictable. Raindrop is a great solution to this.
Hyped for the launch, been alpha-testing Raindrop and it's legit. Congrats on the launch folks <3
@benhylak and the team are experts in AI products!
this is a real problem and I'll try this out in Orango AI.
btw the autocomplete in the "What does your AI product do?" onboarding step was sooo good 🫡
we at new.computer have been users of raindrop since our launch - they've been invaluable for helping us figure out both the macro picture - what's actually going on with user behavior in our product - & tracking micro behaviors / regressions we're trying to fix & eliminate. sentry for natural language AI is long overdue & this team is the right one to build it!
This is going to be great not just for product folks, but also for AI engineers who will start with a much cleaner dataset for fine-tuning
Looooove the rebrand. Also nobody is doing LLM observability like this. Total gap in the market and Raindrop is filling it. Nice work.
Congratulations!
I always worry if my AI agent gives a strange offer to my client. I tried to make GPT not to do those kind of things, but now I realized that it's better managing the issues quickly than making AI perfect.
