AI for Making Podcast
Interviews, insights, and real-world strategies for building AI-powered applications. Deep dives with industry experts who've shipped AI products at scale.
Solving the 70% Problem: Why AI Code Fails in Production
Sarah Chen from Cursor joins us to discuss the infamous '70% problem' - why AI-generated code works great in demos but fails in production. We dive deep into debugging strategies, testing patterns, and how to bridge the gap between prototype and production-ready AI applications.
AI Art Revolution: From Midjourney to Production
Marcus Rodriguez, Creative Director at NVIDIA's AI Research Lab, shares insights on the state of AI art tools in 2025. We discuss Midjourney v7, Flux Pro, Ideogram 2, and how creative professionals are integrating AI into their workflows while maintaining artistic integrity.
Building AI Music Tools That Musicians Actually Want
Anna Chen, founder of AudioCraft AI, discusses the technical and creative challenges of building AI music generation tools. From Suno vs Udio comparisons to the future of AI-human collaboration in music production, we cover what's working and what's not in AI music.
The Real Cost of AI Video Production
David Park from RunwayML breaks down the economics of AI video generation. We explore cost per minute, quality trade-offs, and when AI video makes business sense vs traditional production. Plus: exclusive insights on upcoming Runway features.
Why AI Agents Fail: Lessons from 1000+ Deployments
Rachel Kim, AI Infrastructure Lead at Anthropic, reveals the brutal truth about AI agent deployments. With data from over 1000 production implementations, we discuss the 88% failure rate, common pitfalls, and what actually works in enterprise AI agent systems.
Fine-Tuning ROI: When Custom Models Make Sense
James Liu, ML Engineering Manager at OpenAI, shares hard data on fine-tuning economics. We analyze when custom models deliver ROI vs. prompt engineering, the hidden costs everyone misses, and emerging alternatives to traditional fine-tuning.
Voice AI That Doesn't Sound Like a Robot
Sofia Martinez, Voice AI Researcher at ElevenLabs, explains the technical breakthroughs making voice AI indistinguishable from human speech. We cover emotional modeling, accent preservation, and the ethical implications of ultra-realistic voice cloning.
No-Code AI: When It Works (And When It Doesn't)
Michael Chen, founder of Zapier's AI division, provides a realistic assessment of no-code AI platforms. We discuss success rates, limitation patterns, and when businesses should graduate from no-code to custom development.
ChatGPT o1: Reasoning Models in Production
Dr. Elena Vasquez, Research Scientist at OpenAI, explains the technical architecture behind reasoning models like o1. We explore chain-of-thought at scale, when reasoning models outperform traditional LLMs, and practical implementation strategies.
Creative AI
AI art, music, video generation and creative workflows
Production AI
Deployment, scaling, and real-world implementation
AI Agents
Building autonomous AI systems and agent architectures
Business AI
ROI, strategy, and enterprise AI implementation
About the Show
Hosted by AI practitioners who've built and shipped real AI products. We focus on practical insights, honest discussions about what works (and what doesn't), and real-world data from production AI systems.
Each episode features industry experts sharing their experiences building AI tools, deploying models at scale, and solving the messy problems that come with putting AI in production. No marketing fluff - just honest conversations about the reality of building with AI in 2025.
Never Miss an Episode
Get episode summaries, guest insights, and exclusive content delivered to your inbox. Plus: early access to new episodes and bonus content.
Subscribe to Newsletter