Velvet

We're excited to announce that Velvet has been acquired by Arize AI, the enterprise platform for AI evaluation and observability. Arize helps organizations ensure AI systems work reliably in production through advanced testing, model monitoring, evaluation frameworks, and debugging tools. We’re joining forces to bring Velvet’s focus on developer experience to the Arize platform. We invite you to use Arize AI for future AI development. Reach out if you'd like help getting set up! https://lnkd.in/eUW7Ves4

We love building on top of Cloudflare. Their Workers, Queues, and KV products make a huge impact on scaling our developer experience. Read more about our architecture on the Cloudflare developer blog - https://lnkd.in/eVBcpiez

Today we’re excited to announce Velvet’s new evaluation features. LLMs are powerful, but inherently unpredictable. Run experiments to decide which models to use, and continuous monitoring to maintain confidence in AI features over time. Watch a quick demo to learn how it works! https://lnkd.in/dtkmHaeb

Want to run ongoing tests of your LLM-powered features? Try out our latest evaluations feature, ready to try this week. https://lnkd.in/e8_8gW_T

Read my interview with Indie Hackers about Velvet's journey navigating product-market fit. The perils of operating in a disruptor's market, 0-1 growth tactics, and how to maintain momentum ↗ https://lnkd.in/e6HjKgtt

Gain confidence that your LLM-powered features will work the way you expect them to. Use our latest Evaluations feature to run experiments across models, settings, and metrics in production. Try it out today! https://lnkd.in/gsGcmhvz

Now you can run evaluations with Velvet! Experiment with metrics on top of models, settings, and prompts. And set them as point-in-time or continuous. Email us if you'd like to try it out! https://lnkd.in/e86TWdhK

We're part of Cloudflare's new startup cohort with a great group of fellow infrastructure builders! Follow along -

I made a video with Velvet about how Find AI optimizes AI apps using data. Take a look: https://lnkd.in/er-wTVit

Analyze LLM costs, models, and usage with our text-to-SQL editor. It's free up to 100k requests, or you can connect your own DB. See a demo of how it works: https://lnkd.in/eWjzk6sa