Thanks for using Velvet, @yasser_ahmed ! Glad it's helping you automate disparate joins across sources. Add collaborators to make it even more powerful. Soon you'll be able to visualize queries, set notifications, and publish queries as API endpoints to use in other parts of your product ecosystem.
As someone that has created several in-app subscription apps, I am interested in your approach. We currently use RevenueCat so familiar with the challenge your are tackling.
When will you publish your SDK specs? Beyond native code (SwiftUI and Kotlin), will you also support React Native and other languages as well? Finally, are you incorporating in PayWall page design for A/B testing messages and offers for signup?
I know you are early in the process and any development details you can share would be appreciated.
Hey Sean - Thanks for your feedback! Our product helps you connect all your data to one unified querying interface. Add your databases, third-party tools, and product eventing - then use our AI editor to ask natural language questions and return results (analytics or records). You can turn those queries into API endpoints to use throughout your app. Feel free to email us to get set up! team@usevelvet.com
Warehouse every AI request to your database. Query logs to analyze usage and costs, evaluate models, and generate datasets. Install the proxy with 2 lines of code. It’s free to get started, and you own your data.
Hey Product Hunt 👋 Emma and Chris here from Velvet. We built an AI gateway to warehouse OpenAI and Anthropic requests to your database. Engineers use Velvet logs to analyze usage and costs, evaluate models, and generate datasets.
Our first product was an AI SQL editor. It worked really well, but we had limited insights into what happened between our app and OpenAI. We started storing requests to our database, giving us the ability to query data directly. It gave us so much more control over developing our AI features.
We didn’t think of it as a product until one of our customers (Find AI) asked to use it. We warehoused over 3 million requests for them in the first week, logging up to 1,500 requests per second during their launch. Now we have many startups using Velvet daily for caching, evaluations, and analysis of opaque endpoints like the Batch API.
It's easy to get started! Once set up, we warehouse every request to your database.
Requests are formatted as JSON. You can include custom metadata in the header - like user ID, org ID, model ID, and version ID. This means you can run complex SQL queries unique to your app. Granularly calculate costs, evaluate new models, or identify datasets for fine-tuning.
We'd love to hear from you! Email team@usevelvet.com with any feedback or questions.
@emmalawler24 Congratulations on the launch! 🎉 I’m curious, how does Velvet handle data security and compliance with such high request volumes? Excited to see how it evolves!
Hey @devindsgbyq - thanks! For teams with large volumes of data in production, we warehouse requests to their own PostgreSQL instance. This resolves most security and compliance concerns since the company maintains full control of data. Our app and infrastructure is built with Cloudflare, Supabase, Neon, Vercel, and OpenAI. All are SOC 2 Type II compliant.
Congratulations on launch! Our company Find AI has been an early customer of the AI gateway, and it's been a game-changer for our engineering team to interact with OpenAI.
Here's some of the things we do with the millions of requests we warehouse weekly with Velvet:
- Cost analysis: What's AI spend per user, per search, for logged out users, and in prod vs. development?
- Debugging: When a query does something weird, trace it back to the OpenAI call. We traced one query that occasionally returned gibberish, and realized that the temperature was just too high.
- Caching: particularly use this on smoke tests, so that we can actually query OpenAI if our prompt/model/etc changes to verify expected behavior, but skip paying money if things are the same
- Testing model upgrades: This week we've been exporting data of some models in gpt-4o, then re-running the exact queries against gpt-4o-mini. To our surprise, a couple of our classification queries had 100% the same accuracy on the mini model, so we could immediately switch and save money.
@philipithomas - thanks! Find AI has been a pivotal early design partner. Their engineering team uses Velvet daily, providing feedback on exactly what they need implemented to run and optimize a high-usage AI application. It's inspiring to see the Find AI product get better every day!
Congrats for the launch and thank you, we are happy users of Velvet, it's not only very simple to set up and use but also highly reliable and very well designed. Kudos team 🙌
@mehdidjabri - it's been great to see the revo.pm product evolve and scale. Excited to be part of your journey as you automate more product management workflows.
Velvet