

Machine Learning Guide
OCDevel
Machine learning audio course, teaching the fundamentals of machine learning and artificial intelligence. It covers intuition, models (shallow and deep), math, languages, frameworks, etc. Where your other ML resources provide the trees, I provide the forest. Consider MLG your syllabus, with highly-curated resources for each episode's details at ocdevel.com. Audio is a great supplement during exercise, commute, chores, etc.
Episodes
Mentioned books

Jul 14, 2025 • 1h 12min
MLA 027 AI Video End-to-End Workflow
Discover how to maintain character and style consistency in AI-generated videos. Explore Google Veo 3's 'High-Quality Chaining' for prosumers and indie filmmakers combining Midjourney, Kling, and Runway for narrative coherence. Learn about using layered ComfyUI pipelines for professional studios and effective tools like Suno for music and ElevenLabs for voice synthesis. Unveil tips for crafting viral short videos, from image concepts to audio editing, tailored for every level of creator.

Jul 12, 2025 • 41min
MLA 026 AI Video Generation: Veo 3 vs Sora, Kling, Runway, Stable Video Diffusion
Explore the cutting-edge of AI video generation with insights on Google Veo's stunning 4K photorealism and integrated audio, setting the standard for the market. Discover OpenAI Sora's storytelling prowess with innovative editing features and Kuaishou Kling's incredible image-to-video motion. The discussion dives into the pros and cons of these tools, emphasizing prompt engineering for optimal results. Plus, a glimpse into the future reveals emerging technologies and aggregator platforms that could transform video creation.

Jul 9, 2025 • 59min
MLA 025 AI Image Generation: Midjourney vs Stable Diffusion, GPT-4o, Imagen & Firefly
The podcast dives into the fascinating split in the AI image generation market. Midjourney shines with high-quality art but struggles with text precision. In contrast, tools like GPT-4o excel in conversational control, making them ideal for marketing and design. Adobe Firefly stands out for its commercial safety with licensed training data, while Stable Diffusion offers users complete control. The discussion also covers effective workflows and the importance of prompt engineering for multimedia production, providing insightful comparisons of these powerful tools.

63 snips
May 30, 2025 • 1h 6min
MLG 036 Autoencoders
T.J. Wilder, a machine learning engineer at Intrepio specializing in generative AI for healthcare, discusses the fascinating world of autoencoders. He explains how these neural networks compress data to improve performance through dimensionality reduction and synthetic data generation. Wilder dives into various types of autoencoders, like variational and sparse, and their applications in enhancing model efficiency and interpretability. He also highlights their significance in addressing healthcare data challenges by generating realistic synthetic data.

79 snips
May 8, 2025 • 45min
MLG 035 Large Language Models 2
Dive into the fascinating world of large language models and their ability to learn from examples without the need for updates. Discover how these models utilize Retrieval Augmented Generation for real-time factual lookups. Explore the rise of autonomous LLM agents that can plan, act, and use tools with persistent memory. Learn about the importance of clarity in prompts and effective prompt engineering techniques that enhance performance. Uncover the exciting benchmarks evaluating their capabilities in STEM reasoning and multimodal tasks.

42 snips
May 7, 2025 • 51min
MLG 034 Large Language Models 1
Discover the fascinating advancements in Large Language Models and the game-changing impact of transformers. Learn how scaling laws reveal the relationship between model size, data, and compute, leading to emergent abilities like in-context learning and multi-step reasoning. Delve into optimization strategies, including the Mixture of Experts architecture and reinforcement learning, which align outputs with human values. Explore the art of prompt engineering and chain-of-thought techniques that enhance accuracy and elevate performance for complex tasks.

50 snips
Apr 13, 2025 • 44min
MLA 024 Code AI MCP Servers, ML Engineering
Discover how Vibe Coding agents enhance the coding process with advanced file management and command execution. Learn about the Model Context Protocol, which improves AI tool interoperability, making automation more efficient. Dive into automating machine learning with AWS and Retrieval Augmented Generation techniques. Explore how CodeAI tools optimize workflows for data cleansing and model management. Plus, find strategies for creating custom metrics and enhancing productivity with Jupyter notebooks.

25 snips
Apr 13, 2025 • 38min
MLA 023 Code AI Models & Modes
Explore the latest in AI coding models, including the impressive Gemini 2.5 Pro and Claude 3.7, highlighting their operational modes and performance. Discover the balance between cost and efficiency, especially for complex tasks. Learn about local vs. hosted models, privacy benefits, and customization options to fine-tune your coding projects. Enhance your coding workflow by integrating AI with Git techniques, and manage tasks efficiently with smart multi-window strategies. Get tips on maximizing your AI tools for seamless coding!

62 snips
Feb 9, 2025 • 55min
MLA 022 Code AI: Cursor, Cline, Roo, Aider, Copilot, Windsurf
Discover the innovative world of 'Vibe Coding,' where AI tools revolutionize software development. Learn about Roo Code and its impressive capabilities, especially its customizable workflows for efficient coding. The discussion also tackles the hype surrounding AI tools, drawing parallels to the cryptocurrency boom. Explore the rise of various code AI tools, their features, and why certain options are gaining popularity while others fall out of favor. Perfect for developers wanting to navigate this dynamic landscape!

38 snips
Feb 9, 2025 • 43min
MLG 033 Transformers
Discover how transformers revolutionized machine learning by replacing outdated recurrent neural networks. The power of self-attention enables massive parallel processing, making them highly efficient. Delve into the mechanics of multi-headed attention, where diverse relationships are captured simultaneously. Positional encodings ensure that the sequence order is maintained. Plus, learn how innovative practices like walking desks can enhance focus while studying complex topics. It's a fascinating exploration of technology and productivity!