OpenArt

OpenArt is where ideas become visual stories. We are a well-known AI platform specializing in image and video generation, offering unique features you can't find in other AI products. We recently launched OpenArt Stories that empowers users to turn any idea into captivating visual stories within minutes.

OpenArt's Technical Stack: Lean and AI-Native, powered by Upstash

We are a truly lean AI-native startup. With just a handful of engineers, we've been using Upstash Redis for quite some time, which has been crucial in helping our small but mighty engineering team develop and deploy Redis-based solutions very productively.

Recently, we launched OpenArt One-Click Story and leveraged Upstash Workflows to support our task processing pipeline. This lean approach aligns perfectly with the industry trend toward AI-native companies that can achieve massive scale with minimal resources. Looking ahead, we're excited about potentially building agentic AI systems on top of Upstash's infrastructure.

What were you using before Upstash?

We previously used GCP Memorystore for Redis since many of our services were running on Google Cloud. However, we encountered challenges with pricing and debugging experiences, which led us to try Upstash after hearing great things about it for some time.

Before launching OpenArt Stories, we had also built internal task scheduling and queue systems for different features, but these required significant maintenance overhead.

What changed after adopting Upstash?

During our implementation of Upstash Workflows, the Upstash team was incredibly responsive to our questions and addressed issues quickly.

Pingren

With Upstash, both our DevOps and developer experience became significantly better than what we experienced with traditional cloud service providers. The observability improvements alone were game-changing for our small team.

Pingren, Founding Software Engineer

This combination of better DevOps, superior DX, and exceptional team support made our migration smooth and, most importantly, allowed us to focus on building innovative AI features rather than managing traditional infrastructure.