Hanselminutes with Scott Hanselman

Latest Episodes:

APIs as the Execution Layer of AI with Postman's Rodric Rabbah Jul 10, 2025

Show notes

In this episode of Hanselminutes, Scott Hanselman chats with Roderick Rabah, Head of Product at Postman Flows, about the evolution of software development, the intersection of APIs and AI, and finding the "right layer of abstraction" for problem-solving. Drawing on his deep expertise in compiler optimization, distributed systems, and serverless computing, Rabah shares his perspectives on building tools that empower developers to create efficiently and explores the paradigm shift toward visual programming and AI-driven automation.

The conversation dives into how Postman is innovating in the software space, how approaches to software engineering are transforming with generative AI, and why embracing new ways of working is critical for staying ahead in this rapidly evolving technological landscape.

Key Topics

[01:08] Introduction of Roderick Rabah: From research scientist to API innovator
[02:14] Evolution of software development: From FPGAs to serverless computing
[03:23] APIs and AI: The transformative intersection powering workflows
[05:33] The rise of tool-calling and agents: Simplifying backend tasks
[07:33] Managing complexity: Why structured APIs make integration seamless
[12:08] Visual programming languages: The paradigm shift for developers
[16:42] Postman Flows: Building applications through visual workflows
[20:24] Embracing generative AI: How senior and junior engineers benefit
[29:02] Deploying with WebAssembly: Making cloud integration accessible
[30:33] Reflections on the future of technology and its impact on software careers

Main Takeaways

API + AI Integration: APIs combined with large language models are unlocking new capabilities for software development by abstracting complex operations and enabling automation.
Visual Programming Paradigm Shift: Applications are increasingly built using visual workflows where developers focus on intent rather than low-level code implementation, driving efficiency and accessibility.
Generative AI Empowerment: Generative AI tools are accelerating the pace of innovation, empowering engineers to fix bugs, streamline workflows, and manage edge cases efficiently.
Structured APIs Critical for AI: Thoughtfully designed APIs with proper documentation and safeguards are essential to ensure that autonomous AI agents interact correctly and securely.
Accessible Deployment: New runtime frameworks, like serverless with WebAssembly, make it easier for developers to deploy applications across the cloud, enabling broader adoption of AI-driven solutions.

Notable Quotes

"Serverless is where you think about servers less." – Scott Hanselman

"At what point does communicating your intent to AI become programming again?" – Roderick Rabah

"Visual programming resonates with builders because it matches the mental model of decomposing problems." – Roderick Rabah

"Technology transforms rapidly. You have to figure out how to wield this immense power." – Roderick Rabah

"Don’t throw away your critical thinking just because AI makes building faster." – Roderick Rabah

Resources Mentioned

Postman Flows – Tools for visual programming and API integrations: postman.com
Replit – Generative coding platform for automating development tasks: replit.com
WebAssembly – Runtime framework for deploying serverless applications: webassembly.org
Books on Compiler Theory: Suggested resource for expanding understanding of abstractions

Follow along for more insights, tips, and conversations with industry leaders. These show notes summarize key moments in the podcast for easy reference and understanding - these show notes were generated by a custom gpt-4o-nano model trained in previous episodes of Hanselminutes

Saving State while Still Growing with Temporal's Preeti Somal Jul 03, 2025

Show notes

In this episode, Scott chats with Preeti Somal, Senior Vice President of Engineering at Temporal, to explore how teams at OpenAI, Stripe, Netflix, and beyond are building long-running, crash-proof applications using Temporal's open‑source durable execution engine. Drawing on her leadership roles at HashiCorp, Yahoo!, and VMware, Preeti breaks down the orchestration challenges in today’s AI‑powered agentic architectures, shares how platform engineering, culture, and developer experience interact, and explains the feedback loops that drive platform improvement. She also offers a dive into resilience patterns like retries, state management, and sagas, and shares lessons on scaling engineering organizations through rapid growth.

SKY ENGINE AI's Dr. Malc Souter on AI-Generated Training Data Jun 26, 2025

Show notes

In this episode Scott talks with Dr. Malc Souter, a computer graphics algorithms PhD and former Hollywood special effects artist, now bringing his rendering expertise to AI at SKY ENGINE AI. They dive deep into the surprising power of synthetic data, exploring when fake can outperform real in areas like medical imaging, defense, and self-driving cars. Malc shares lessons from the visual effects trenches, discusses how custom rendering engines are reshaping machine learning pipelines, and unpacks the nuanced tension between privacy and progress in the age of computer vision.

https://www.skyengine.ai

Dave Barry is the Class Clown: The Memoirs of a Professional Wiseass Jun 19, 2025

Show notes

When authors want to talk about their book they go on a book tour and often they go to large companies like Microsoft and speak to us in person. I had the opportunity to interview Dave Berry in person and I jumped at it. I grew up reading his humor column syndicated in The Oregonian, and his brand of sarcasm and dry humor matches mine. This episode is that raw recording in association with and with the permission of Microsoft's Outside in Lecture Series program. He was a lovely gentleman and we enjoyed spending time together, so if it seems like we're picking on each other, we are.

Dave Barry's Class Clown: The Memoirs of a Professional Wiseass

Startup Mindsets with Earl Valencia Jun 12, 2025

Show notes

What does it really take to succeed in today’s fast-paced, innovation-driven world? In this episode of Hanselminutes, Scott Hanselman chats with Earl Valencia—acclaimed tech leader, venture-backed founder, and author of Startup Mindsets: A Blueprint to Thrive in an Innovation-Driven and Globally Connected World. Earl brings real-world insights from Silicon Valley, Southeast Asia, and startup scenes around the globe. He talks about the habits, mindsets, and leadership styles that help founders stand out—not just another success story, but a practical guide for anyone trying to build something meaningful. Whether you’re launching a startup, working in tech, or just curious about how big ideas come to life, this conversation will leave you inspired to think bigger and bolder.

https://www.startupmindsets.com/book

AI and visualizing multidimensional vectors with Pamela Fox Jun 05, 2025

Show notes

Ever wonder how AI “sees” high-dimensional data? In this special 1,000th episode of Hanselminutes, Scott Hanselman chats with AI expert Pamela Fox about the fascinating world of visualizing complex data. Pamela breaks down how those abstract numbers and vectors can be turned into visuals we can actually understand—basically helping us see how AI thinks. They also talk about the challenge of explaining advanced AI stuff in plain English, and why it matters for devs, data folks, or really anyone curious about what’s going on behind the scenes. If you've ever wanted a peek into the brain of an AI, this is a fun and eye-opening place to start.

AI Code Reviews with CodeRabbit's Howon Lee May 29, 2025

Show notes

What if AI could not only speed up your code reviews—but actually make them fairer? In this episode of Hanselminutes, Scott Hanselman chats with Howon Lee, Senior Software Engineer at CodeRabbit, about how AI is changing the game when it comes to code reviews. They dig into the unspoken dynamics that can pop up in traditional reviews—like how seniority or team politics sometimes matter more than the code itself. Howon shares why AI can be a helpful, unbiased voice that gives everyone clear, consistent feedback, no matter their experience level. They also talk about how tools like CodeRabbit are working to make software development more inclusive and collaborative.

http://coderabbit.ai

Azure AI Foundry (from BUILD 2025) with Yina Arenas May 22, 2025

Show notes

Learn how to supercharge your AI development with our integrated signals loop that connects model choice, knowledge retrieval, fine-tuning, orchestration and memory —anchored by observability and trust. Design, customize, and manage intelligent agents using open standards and protocols such as Model Context Protocol (MCP) and Agent-to-Agent (A2A) to connect with tools and drive collaboration. Gain insights into advanced orchestration, tracing, and monitoring to streamline decision-making, boost efficiency, accelerate time to market, and lower costs.

NOTE - This is a bit of a meta-episode, recorded live at Microsoft Build, this is a discussion about how Scott might use AI to help produce his podcast!

Azure AI Foundry: The AI app and Agent Factory | BRK155

Is AI what Robotics needs? with Jasmine Lawrence Campbell May 16, 2025

Show notes

In this episode of Hanselminutes, Scott sits down with AI/ML & Robotics leader Jasmine Lawrence Campbell to explore the evolving relationship between artificial intelligence and robotics. Jasmine shares insights on how AI is shaping the future of robotics, the challenges of integrating AI into robotic systems, and why she believes there's a place for everyone in AI. Whether you're a developer, researcher, or simply curious about the intersection of AI and robotics, this conversation offers fresh perspectives on the technology driving automation forward.

https://www.jasminelawrence.com

Is Vibe Coding Real? with James Montemagno May 08, 2025

Show notes

Vibe Coding has folks talking and "vibing entire applications." But is it valid? Should one use AI agents to create apps that go directly into production, or is it just appropriate for prototyping? Scott talks to James Montemagno who recently vibed a 17,000 line application and only wrote 20 bespoke lines himself. Is this the future of programming or did James get lucky? Scott takes the sceptical view in this spicy episode.

Our TOPPODCAST Picks

Follow Us

Stay Connected

Hanselminutes with Scott Hanselman

Key Topics

Main Takeaways

Notable Quotes

Resources Mentioned

Related Podcasts

Links

Stay Connected