Hanselminutes with Scott Hanselman

Trading Engineering Durability for Short Term Vibe Coding Speed

Vibe coding sacrifices the long-term durability of software for immediate speed, which results in fragile systems that are difficult to maintain. Master the fundamental mechanics of engineering to build lasting software and gain a competitive advantage in an era of mediocrity.

Prioritizing Reliability and Mass Over Complexity in Defense Innovation

In hostile environments, complexity is a liability. You can master the defense market by prioritizing modularity and simple, attritable designs that survive failure, rather than over-optimizing for ideal conditions that rarely exist in the field.

Demystifying AI Mechanics for Durable and Predictable Systems

Treating AI like a mystical oracle creates fragile systems and builds up technical debt. You should master the mechanics of tokenization and weight evolution to build predictable, durable, and reliable production tools.

Securing Developer Workstations Through Deterministic AI Engineering Standards

Developer workstations have become the main entry point for enterprise security breaches. By building deterministic security checks directly into the AI coding workflow, teams can stop vulnerabilities before they ever reach a pull request.

Stack Overflow Shifts to Verified Knowledge for AI Agents

Stack Overflow is becoming a verification layer for AI agents. By grounding agent workflows in human-verified, traceable knowledge, teams can stop hallucinations and replace unreliable AI output with code ready for production.

Prioritizing Foundational Infrastructure Over Performative AI Deployment

AI innovation in the Global Majority often fails because it prioritizes performative tools over foundational infrastructure. Build long-term value by shifting focus from rapid deployment to sovereign, locally-maintained systems and resilient capacity-building.

Transitioning From Snapshot-Based Versioning to Event-Sourced Collaboration

Software development is stuck in a snapshot model that overlooks how people actually work together. By moving from static commits to event-sourced versioning, we can capture the hidden history of development and turn code into a transparent, collaborative canvas.

Advancing Accessibility Through Multiline Spatial Data Hardware

Linear, single-line Braille displays force users to mentally reconstruct complex data, which creates a cognitive tax. Multiline, spatial hardware and data-first design offer a more robust interface that benefits every user.

Observability Strategies to Mitigate AI-Induced Operational Debt

AI-generated code speeds up development, but it also creates a dangerous feedback void. Without robust observability, teams simply accumulate operational debt. This leaves them unable to debug their systems or develop the human judgment required for long-term resilience.

Cloud Commitments: Balancing Cost Savings With Financial Risk

Cloud commitments offer deep discounts but create significant financial risk. Secure savings and flexibility with insured commitments that hedge against uncertainty.

Leveraging Mainframe Data Gravity for Modern Enterprise Infrastructure

Mainframe modernization succeeds when teams focus on developer experience and data gravity rather than just moving languages. By building modern CI/CD pipelines into high-resiliency architecture, organizations transform legacy systems into agile, high-performance private clouds.

Managing AI Agent Cognitive Load for Developer Flow

AI agents introduce immense cognitive load, demanding developers master expressing judgment and taste amidst generated output to avoid overwhelming chaos and maintain flow state.

Resilience Engineering: Designing Systems for Failure Not Prevention

Security is an illusion; true robustness lies in designing systems for inevitable failure and rapid recovery, not just prevention.

AI-Driven Introspection Risks Stunting Social Skills and Novel Problem-Solving

AI can create an echo chamber, stunting growth and problem-solving. Cultivate diverse external inputs to gain adaptability and genuine understanding.

Early Internet's Utopian Vision Eroded by Systemic Complexity

The internet's utopian vision fractured as early design choices, prioritizing flexibility, created unforeseen societal complexities and an "open adversarial system."

Persistent AI Memory Accelerates Scientific Discovery Beyond Stateless Models

AI's true breakthrough isn't just agents, but persistent memory. Learn how storing past research, even incomplete thoughts, prevents reinventing the wheel and unlocks exponential gains in discovery.

Agentic Workflows Amplify Developer Discipline for Software Quality

AI amplifies developer skill, not mediocrity. Harnessing its power demands rigorous discipline, strategic guidance, and intentional design to build better software.

Human Expertise Essential for Robust AI Inference Engineering

AI's rapid content generation can erode trust. True expertise in AI inference demands deep human effort, not shortcuts, for lasting advantage.

Legacy Abstractions Hinder AI Innovation and Performance

Legacy programming languages designed for 2010 hardware now limit AI innovation and create competitive disadvantages. Embrace unified models for precise control over modern silicon.

Agentic AI Requires "Good Taste" and User Education

AI agents are evolving beyond prompts to deeply integrated systems that extend your agency, redefining control and data ownership. Understand the "ambiguity loop" and "good taste" to build the next generation of intelligent applications.

AI Productivity's Human Cost--Developer Exhaustion and Wealth Extraction

AI-driven productivity paradoxically drains developers, creating burnout and questioning the value of relentless output. Discover the "AI Vampire" phenomenon and its implications for sustainable work.

AI Enhances Code Reviews With Contextual Feedback

AI transforms code reviews from superficial checks to contextual, collaborative feedback, fostering a healthier developer culture and a significant competitive advantage.

AI Agent Sandboxing: Controlled Risk Over False Safety

AI agents create a false sense of security; understand the nuanced differences between traditional isolation and agent sandboxing to avoid costly mistakes and gain a strategic advantage.

AI Empowers Personal Software Development and Hacker Mentality

AI empowers anyone to build custom software in minutes, replacing expensive SaaS and sparking a new era of personal creation and problem-solving.

Fabulous Adventures: Deep Understanding Beyond AI Coding

Master complex algorithms and data structures beyond typical curricula to translate deep understanding into elegant code, not just AI-generated syntax.

Game Development Principles Yield Performant, Lightweight Windows File Explorer

Experience game-like rendering speeds and a tiny 2-3MB footprint in a Windows file explorer built with custom, low-level techniques.

Zig's Explicitness and C Interoperability Revolutionize Systems Programming

Zig offers a modern, explicit approach to systems programming, simplifying C interoperability and empowering developers with compile-time code execution.

Agentic AI Security Risks: Expanding Attack Surface and Unpredictable Behavior

Agentic AI's expanding attack surface and "jagged intelligence" create unprecedented security risks, demanding provable guarantees for trustworthy, critical decision-making.

Amplifying Human Agency Through Intentional AI Design

AI amplifies human agency when designed to augment capabilities, enabling delegation of computation to machines while humans focus on complex problem-solving and innovation.

Esoteric Programming Languages: Art, Absurdity, and Computational Exploration

Esoteric programming languages push code beyond utility, exploring art, absurdity, and conceptual design with extreme constraints that reveal profound computational insights.

Airport IT: The Unseen City Beneath the Runway

Airport IT leadership balances stability and innovation, managing a vast physical and digital infrastructure akin to a mini-city, where seamless operation is the ultimate success.

C++: Powering Modern Infrastructure, Not Just Old Code

C++ remains vital for cutting-edge tech, powering GPUs and AI; its perceived age masks continuous evolution, prioritizing robust, maintainable engineering over fleeting trends.

Postgres: Four Decades of Open-Source Extensible Dominance

Postgres' 40-year success hinges on its extensible design, open-source community, and collaborative, vendor-independent development, making it a robust, adaptable database for every scale.

Running History: Why Vintage Computers Matter Now

Industry Experience Elevates Academic Theory for Engineers

Modernize .NET Apps: AI, Cloud, and Testing Accelerate Migration