Snowglobe is praised for its ability to simulate real-world user behavior, allowing LLM teams to identify and address edge cases before launch. Users highlight its intuitive design and engaging experience, noting the tool's capacity to make testing workflows smoother and more reliable. The attention to detail and creativity in its execution leave a lasting impression, making it a valuable asset for designers and innovators. Overall, Snowglobe is seen as an essential tool for enhancing creativity and ensuring safer production rollouts.
Snowglobe
🫡 Hi Hunters, I’m Shreya, co-founder of Snowglobe (by Guardrails)!
If you’ve built AI agents, you know how challenging it is to test them. How do you even begin formulating a test plan for a technology whose input space is infinite?
Most teams fall back on a small ‘golden’ dataset, maybe 50 to 100 hand-picked examples.
It takes ages to put together, and even then, it only covers the happy paths, missing the messy reality of real users.
That’s how you end up with an agent that’s perfect in development, but starts hallucinating, going off-topic, or breaking policies as soon as it meets real-world scenarios.
🔮 Snowglobe fixes this problem by creating a high fidelity simulation engine for conversational AI agents!
We build realistic personas and run them through thousands of simulated conversations with your AI agent BEFORE you go to production.
Our customers are already generating tens of thousands of simulated conversations with Snowglobe, allowing them to speed up what used to take weeks of manual scenario hand-crafting and catch potential issues before your users do.
🦾 Why we built this
Before Snowglobe, I spent years building self-driving cars at Apple. Weirdly enough, he challenge with LLMs is surprisingly similar to self-driving cars: huge input space, and high stakes when something fails.
In that world, we used high fidelity simulation engines to test cars in even the most risky, and rare scenarios.
Waymo, for example, logged 20+ million miles on real roads, but over 20+ BILLION in simulation before launch.
Our goal was simple - bring that same simulation-first mindset to AI agents, starting with AI chatbots.
💅 How we’re different
→ We perform rich persona modeling to ensure a lot realism and diversity in the scenarios we generate. This is not the same as asking ChatGPT to generate synthetic data for you, which then all sounds like the same ChatGPT-voice created them.
→ We simulate full conversations, not just one-off prompts.
→ We ground scenarios in your agent’s context so they’re relevant to your use case.
→ Unlike conventional redteaming tools, we test normal-user behaviors as well as edge cases.
→ You can export scenarios straight to 🤗 Hugging Face or your favorite eval & tracing tool.
🫶 Who is this for?
If you’re building a conversational AI agent and:
→ You’re stuck testing with a tiny dataset,
→ You want to create a dataset for finetuning your LLM, or
→ You’re spending too much time creating test sets manually, or
→ You want to run QA or pentesting before launch, or
You should give Snowglobe a try.
💸 Snowglobe.so is live and ready for use!
Product Hunt fam gets $25 worth of free simulated conversation generation with the code PH25 (in addition to the $25 worth of free credits on start).
We’ve poured a lot of engineering, research and, most importantly, love into this project.
We’re excited to see how we can help you test your chatbots better! 🙌
CoSupport AI
@shreya_rajpal very cool approach 👌
testing agents before they go live is huge - especially when real users have zero tolerance for hallucinations or off-brand replies. always curious to see how others tackle the “make it production-ready” challenge.
@shreya_rajpal Love the “pre-users before users” idea. Do you support persona libraries and replayable sessions for regressions?
Snowglobe
@shreya_rajpal @anwarlaksir Something we've been thinking about! Right now, we have a number of personas that can be used across simulations for the same app, but it's not as organized as a library (yet)
@shreya_rajpal @anwarlaksir @zayd_simjee would love to contribute to this product, sounds exciting!
This is incredible! I've worked on similar tools for DeepMind & YouTube.
How much control do users have over personas? Can I effectively write 'system prompts' to configure different personas & then have them test AI agents?
For example, I want to make my agent safe to deploy to teens – want to create a number of teen & teen-risk related personas to red team the agent. From the website/demo this doesn't seem to be configurable at the moment.
Love this launch & direction!
Snowglobe
@devansh_tandon1 Hey Devansh! So we put a lot of work in crafting the personas for you, what you can do is this:
Add a Custom Risk: For this you can add your prompt to flag any conversation or message that you don't want your agent to respond with
You can set your simulation intent for what bad cases you want to catch
Based on that we will generate effective personas to catch those cases. Maybe give it a shot, and let us know if the personas end up working well as is
Ito
Snowglobe
@barroncaster thanks!
You can get started with as little as a description of what your app does. As you feed it more information, the better the personas get.
Ito
Reflex
Looks awesome! Im developing coding agents for ai app building and we have evals but they don’t always match user inputs. Snowglobe sounds perfect for making our test cases more realistic, can’t wait to try it out
Snowglobe
@picklelo Excited for you to try it out!
stagewise
That sounds quite useful for basic agent scenarios, congrats on the launch!
I guess it's not suited for complex agents running on a custom backend with multiple tools? @shreya_rajpal
Super Intern
How customizable are the personas? For example, can I tweak them to simulate niche user groups, like non-native English speakers?
Super Intern
Valuable for me! It's hard ro find early-stage users! And the feedback is so important!
@shreya_rajpal Do I need to define the persona of user, or it will auto generate for me accourding to my product?