RightNow AI v2 CUDA kernel optimizer with real-time profiling

After a ton of feedback, I've pivoted RightNow AI into v2: a browser-based chat interface that helps you write high-performance CUDA code without touching NSight or spending hours profiling.

Just tell it what you're trying to do, and it generates optimized CUDA kernels—sometimes outperforming NVIDIA’s own. We’re focusing heavily on real-time bottleneck profiling to keep things fast and smart.

Try it out here: https://www.rightnowai.co/ Also on Product Hunt: https://www.producthunt.com/posts/rightnow-ai-2

Would love your feedback and support!