Machine Learning into Production podcast, interview style
Episodes
3 Aug 2026
"OpenAI's capacity model and shifting AI pricing strategies highlight financial risks of external LLMs, prompting cost optimization and open-source alternatives for sustainable adoption."
Open episode27 Jul 2026
"SDKs now see hundreds of millions of downloads annually, with a focus on minimal, extensible designs and a major MCP update shifting to stateless protocols for scalability, balancing simplicity with complexity while prioritizing stability and future-proofing."
Open episode20 Jul 2026
"Fast MCP streamlined the Multi-Chat Protocol, dominating the market with simplicity and efficiency, while evolving to support interactive UI apps, Python-based token-efficient interfaces, and addressing security and scalability challenges, with AI tools enhancing personal and professional workflows."
Open episode13 Jul 2026
"Modern software development faces bottlenecks from limited human resources and AI-driven shifts, transforming productivity, SaaS models, and workflows while straining infrastructure and open-source ecosystems."
Open episode6 Jul 2026
Integrating AI agents with enterprise systems via APIs presents security risks from untrusted access, requiring solutions like the Multi-Cloud Protocol, zero-trust models, and GraphQL to balance innovation with safeguards against data exposure and autonomous decision risks.
Open episode6 Jul 2026
Recommended: There is more than one way to build with AI
Advancements in AI tools like Hugging Face MCP and Fast Agent simplify LLM integration for innovative workflows, emphasizing idea-driven development, Rust's performance, open-source models (e.g., Gemma 4, Quen), and accessible tools for non-experts, while balancing efficiency, transparency challenges, and evolving SDKs.
Open episode6 Jul 2026
The Model Communication Protocol (MCP) enables secure AI-to-tool integration via APIs, with DeepL promoting it through global workshops, hackathons, and practical examples like a Python server, emphasizing security, implementation challenges, and hands-on learning to bridge technical gaps and enhance AI workflows.
Open episode6 Jul 2026
The evolution from low-level programming to high-level abstractions, AI-driven natural language coding with its ambiguity and reliability challenges, and the rise of durable execution as a resilience layer for long-running processes highlight ongoing trade-offs between automation, correctness, and infrastructure complexity in software development.
Open episode
3 Jul 2026
Transitioning budget management to developers via AI-driven agentic workflows in service systems, addressing matcha production challenges in Nantou County, language processing complexities, infrastructure limitations, and open-source tools for regional agricultural projects.
Open episode1 Jul 2026
Agentic search systems face challenges in efficiency and contextual understanding, requiring context-aware AI, adaptive evaluation metrics, collaborative frameworks, and optimizations in retrieval, memory, and agent training to improve performance and scalability.
Open episode