Author(s): Shashwat (Shawn) Gupta Originally published on Towards AI. TLDR (from my X.com/LinkedIn post) Recently read MIT’s “State of AI in Business 2025” Report, which reveals a telling disconnect: despite $35–40B invested, 95% of AI pilots fail to deliver business impact, yet …

Author(s): Shashwat (Shawn) Gupta Originally published on Towards AI. Speaker diarization doesn’t always require sophisticated multi-stage neural pipelines to deliver practical results. For many real-world applications, particularly those focused on post-processing clean audio content like podcasts or interviews, a surprisingly simple approach …

Author(s): Shashwat (Shawn) Gupta Originally published on Towards AI. The AI community knows that DeepSeek censors some content. However, in this post, I bring out common approaches to jailbreak the model and get relevant information. The whole idea is to fool the …

Author(s): Shashwat (Shawn) Gupta Originally published on Towards AI. GradCAM is one of the simplest techniques to get explainability insights into model prediction. I was surprised to find that while there are many blogs on Medium about using GradCAM with ResNet, there …