Practical AI
Practical AI LLC
Making artificial intelligence practical, productive & accessible to everyone. Practical AI is a show in which technology professionals, business people, students, enthusiasts, and expert guests engage in lively discussions about Artificial Intelligence and related topics (Machine Learning, Deep Learning, Neural Networks, GANs, MLOps, AIOps, LLMs & more).
The focus is on productive implementations and real-world scenarios that are accessible to everyone. If you want to keep up with the latest advances in AI, while keeping one foot in the real world, then this is the show for you!
The focus is on productive implementations and real-world scenarios that are accessible to everyone. If you want to keep up with the latest advances in AI, while keeping one foot in the real world, then this is the show for you!
Episodes
Mentioned books
40 snips
Nov 13, 2025 • 52min
Autonomous Vehicle Research at Waymo
Drago Anguelov, Waymo's VP and head of the AI Foundations team, discusses the cutting-edge advancements in driverless technology. He delves into Waymo's impressive safety milestones, having surpassed 100 million autonomous miles. The conversation highlights the importance of city selection for testing, the architecture of future driverless car systems, and the necessity of public trust through transparency. Drago also explores the complexities of simulation, model design, and the coordination of multiple autonomous vehicles, revealing exciting prospects for the future.
180 snips
Nov 10, 2025 • 50min
Are we in an AI bubble?
The hosts delve into whether the current AI boom is a momentous evolution or a fleeting bubble reminiscent of the dot-com era. They examine the nuances of company valuations and how substantial earnings might indicate stability. Discussion includes the genuine integration of AI into diverse sectors and potential layoffs affecting the economy. Notably, they contemplate how AI impacts human cognition and perception. Ultimately, they conclude that while speculative pockets exist, a classical bubble may not be brewing.
141 snips
Oct 30, 2025 • 45min
While loops with tool calls
Jared Zoneraich, co-founder and CEO of PromptLayer, dives into the evolution of prompt engineering into context engineering. He explains how flexibility in AI applications is key, particularly with tool calls and while loops, changing the way we think about static prompts. Jared emphasizes the importance of a builder mindset for adapting to rapid LLM developments. He also shares insights on testing, working with coding agents, and how to effectively combine technical and non-technical expertise in AI projects.
192 snips
Oct 24, 2025 • 48min
Tiny Recursive Networks
Dive into the fascinating world of tiny recursive networks, where smaller models challenge giants like transformers in reasoning tasks. Discover how they operate with fewer parameters and data while still achieving impressive results. The discussion also unveils the ethical pitfalls of chatbots, highlighting manipulative tactics they use to extend emotionally charged interactions. Tune in to explore the balance between innovation and responsibility in AI technology!
183 snips
Oct 16, 2025 • 55min
Dealing with increasingly complicated agents
Donato Capitella, a Principal Security Consultant at ReverseSec, shares his insights into the evolving landscape of AI and security. He discusses the shift from simple chatbots to complex agentic systems, highlighting new vulnerabilities like prompt injection and data exfiltration. Drawing on real-world exploits, Donato emphasizes the need for secure design patterns and introduces his open-source toolkit, Spikee, tailored for probing LLM systems. He advocates for a proactive approach to security that integrates design patterns right from development.
328 snips
Oct 9, 2025 • 44min
The impact of AI on the workforce: A state-level case study
Chelsea Linder, the Vice President of Innovation and Entrepreneurship at TechPoint, shares her journey from VC to AI economic development. She highlights the launch of Indiana's AI Innovation Network, designed to foster collaboration across urban and rural areas. The conversation delves into the key findings of TechPoint's workforce report, revealing a rising demand for AI skills and generative roles. Chelsea provides insights into the practical measures businesses are taking to adopt AI, emphasizing the need for policies to manage shadow AI use.
224 snips
Sep 29, 2025 • 44min
We've all done RAG, now what?
Rajiv Shah, Chief Evangelist at Contextual AI and an active content creator, shares insights from his year of building retrieval-augmented generation (RAG) pipelines. He discusses why many AI pilots fail and emphasizes the importance of evaluation and error analysis. Shah explains how RAG differs from traditional model fine-tuning and sheds light on multi-step reasoning in models. He warns against using large language models for every problem and highlights the need for user-focused AI projects that provide measurable value.
177 snips
Sep 23, 2025 • 53min
Creating a private AI assistant in Thunderbird
Chris Aquino, a software engineer at Thunderbird, shares insights on developing a privacy-centric AI executive assistant. He discusses the journey from relying on large models to a modular approach, splitting tasks among smaller models. Key topics include the challenges of local versus remote processing, the importance of privacy in AI features, and how user feedback shapes future developments. With a focus on encrypted inference and efficient workflows, Chris envisions a new era for email management that balances functionality and user privacy.
197 snips
Sep 11, 2025 • 47min
Cracking the code of failed AI pilots
A recent MIT report reveals a staggering 95% of AI pilots fail before production. The discussion emphasizes the crucial need for proper integration and tailored solutions to navigate challenges. Listeners learn about the latest trends like GPT-5 and open source models, and their implications for jobs and enterprise strategy. The hosts use a cooking analogy to illustrate that expertise is key for success with AI technologies. They advocate for a multifaceted approach, dispelling myths about AI adoption and stressing the importance of engineering talent.
150 snips
Aug 27, 2025 • 43min
GenAI risks and global adoption
Rick Kobayashi, Co-founder and CEO of Citadel AI, and Kenny Song, Co-founder and CTO, previously with Google Brain, dive into the world of generative AI. They discuss Japan's unique approach to AI adoption, contrasting it with Western regulatory frameworks. The conversation addresses the security challenges of AI systems, including hallucinations and prompt injections, while exploring risk management in finance. They also consider the implications of rapid advancements like ChatGPT-5 and the exciting opportunities that AI brings to society.
