https://harsidaksinghbanwait.com/blog Essays and build logs by Harsidak Singh Banwait on edge AI, reinforcement learning, wearable hardware and the research behind them. en Sun, 23 Aug 2026 08:27:16 GMT https://harsidaksinghbanwait.com/blog/6 https://harsidaksinghbanwait.com/blog/6 Mon, 27 Apr 2026 00:00:00 GMT Transformers have been wasting capacity for years. Exclusive Self Attention (XSA) is a surgical correction that projects out self-information, improving contextual modeling with zero overhead. https://harsidaksinghbanwait.com/blog/5 https://harsidaksinghbanwait.com/blog/5 Sun, 12 Apr 2026 00:00:00 GMT Exploring the core failure of retrieval-augmented generation, what makes SELF-RAG fundamentally different, and why teaching a model to question itself might be the most important idea in language model research right now. https://harsidaksinghbanwait.com/blog/4 https://harsidaksinghbanwait.com/blog/4 Mon, 22 Jan 2024 00:00:00 GMT An exploration of the intersection between quantum computing and artificial intelligence. Understanding how quantum algorithms could revolutionize machine learning and what it means for the future of AI. https://harsidaksinghbanwait.com/blog/1 https://harsidaksinghbanwait.com/blog/1 Mon, 15 Jan 2024 00:00:00 GMT A deep dive into the challenges and breakthroughs of creating wearable AI technology. From hardware selection to software optimization, here's my journey building smart glasses that can interpret visual environments. https://harsidaksinghbanwait.com/blog/2 https://harsidaksinghbanwait.com/blog/2 Mon, 08 Jan 2024 00:00:00 GMT A comprehensive developer's guide to modern Reinforcement Learning, tracing the evolution from Markov Decision Processes to TRPO, GAE, and Proximal Policy Optimization (PPO). https://harsidaksinghbanwait.com/blog/3 https://harsidaksinghbanwait.com/blog/3 Mon, 01 Jan 2024 00:00:00 GMT Exploring the transformative impact of large language models on software development. From code generation to debugging, LLMs are reshaping the developer experience.