The Practical AI Digest
A generative AI-powered podcast by M Mostagir Bhuiyan. Each episode distills advanced AI/ML topics into real-world insights for practitioners.
About the Podcast
The Practical AI Digest is built using NotebookLM to synthesize research papers, technical reports, and industry developments into conversational episodes. The format is designed for engineers and technical leaders who want to stay current on AI/ML without reading every paper. Episodes cover topics from model architecture and training techniques to deployment patterns and infrastructure decisions.
Hosted by M Mostagir Bhuiyan, the podcast draws on hands-on experience building AI infrastructure, writing research on retrieval-augmented generation and micro-containerization, and leading cloud platform teams.
Listen On
Episodes
Mixture of Experts at the Edge: Running 30B Parameter Models on Your Laptop
July 16, 2026
A 30B parameter model runs on a MacBook because only 3B parameters fire per token. Mixture of Experts splits memory cost...
The Agent Interoperability Problem: Why Your AI Agents Can Not Talk to Each Other
July 2, 2026
90% of enterprises deploy AI agents. Only 23% scale them. The gap is interoperability. Three protocols, MCP, A2A, and AC...
KV Cache Compression: The Memory Wall Nobody Talks About
June 18, 2026
Your GPU is not compute-bound. It is memory-bound. The KV cache is eating half your inference budget, and two ICLR 2026 ...
Context Rot: Why Million-Token Windows Quietly Fail
June 4, 2026
Models advertise million-token windows but accuracy degrades well before the limit. Three recent studies, the mechanisms...
LLMOps: Operating Large Language Models in Production
May 26, 2026
Building an AI model is one thing: keeping a large language model running reliably in the real world is another. In this...
TinyML & Edge AI: Machine Learning on Devices
May 12, 2026
In this episode, we explore how AI is moving from the cloud to tiny devices. TinyML is the field of optimizing models an...