How I AI
By Claire Vo
How I AI, hosted by Claire Vo, is for anyone wondering how to actually use these magical new tools to improve the quality and efficiency of their work. In each episode, guests will share a specific, practical, and impactful way they’ve learned to use AI in their work or life. Expect 30-minute episodes, live screen sharing, and tips/tricks/workflows you can copy immediately. If you want to demystify AI and learn the skills you need to thrive in this new world, this podcast is for you.
59 episodes
All Episodes
Prioritizing Trust and Design for Consumer-Grade AI Agents
Successful AI agents succeed by managing the trust tax instead of just maximizing raw intelligence. By focusing on emotional design and gradual permission flows, developers can turn complex tools into reliable, long term companions.
View Episode Notes →
Building Enterprise AI Through Infrastructure and Governance
Successful enterprise AI is not a product challenge. It is a requirement for governance and infrastructure. Companies gain an advantage by treating AI agents like production code and stabilizing data platforms before they scale.
View Episode Notes →
Autonomous UI Navigation Replaces Custom AI--Native Abstractions
GPT-6 Astra makes the No-UI movement obsolete by allowing systems to navigate complex visual interfaces autonomously. Developers can now use existing SaaS tools rather than building custom abstractions, moving the bottleneck from interface complexity to user ambition.
View Episode Notes →
Prioritizing Agent Durability Over Custom Infrastructure Maintenance
Custom agent stacks often turn into liabilities when the cost of maintaining them outweighs their actual value. Instead of building complex infrastructure, focus on deploying durable, reliable platforms that prioritize long-term stability over technical elegance.
View Episode Notes →
Accelerating Technical Learning Through Impractical Hardware Projects
Build impractical hardware projects to speed up your technical learning. When you prioritize play over utility and use a vibe-coding workflow, you stop worrying about perfection and pick up the integration skills you need for the future of agent-based coordination.
View Episode Notes →
Managing Model Friction to Extract High-Quality AI Output
The most capable AI models are getting harder to use, as they trade raw performance for neurotic, verbose personalities. Master the friction of agent management to extract high-quality output while ignoring the distracting slop.
View Episode Notes →
Transitioning From Chatbot Interaction To Autonomous Digital Labor
AI agents are moving beyond simple chatbots to become autonomous digital employees. By delegating repetitive browser tasks and exhaustive quality testing to these agents, you can remove digital busywork and find hidden friction in your workflows.
View Episode Notes →
Building Competitive Advantage Through Systematic AI--Assisted Content Distribution
Technical advantages fade fast, but distribution is a lasting asset. Turn your team into a marketing engine by using AI to interview them and extract their expertise. This creates scalable content that grows in value over time.
View Episode Notes →
Scaling Productivity Through Persistent Local AI Infrastructure
Move from reactive prompting to ambient intelligence by running local AI agents around the clock. This autonomous infrastructure removes cloud costs and latency, turning your hardware into a high-throughput software factory that operates while you sleep.
View Episode Notes →
Prioritizing Practical Effectiveness Over Theoretical Precision in AI Models
Trying to be theoretically perfect often stalls product development, resulting in code that is brittle and hard to use. Focus on what actually works rather than over-analyzing every detail. This approach helps you build flexible systems that prioritize speed and real value for the user instead of rigid, flawless logic.
View Episode Notes →
Transforming AI Agents Into Reliable Engines Through Deterministic Harnesses
General-purpose AI tools often fail in production environments because they lack structure. You can turn chaotic chatbots into reliable engines by building opinionated harnesses that codify workflows, enforce constraints, and automate high-stakes, repetitive tasks with precision.
View Episode Notes →
Scaling AI Systems Through State-Based Orchestration Instead of Prompting
Stop treating AI as a conversational partner and start managing it as a persistent worker. By replacing fragile chat interfaces with structured state machines, you can build scalable systems that deliver real business value.
View Episode Notes →
Prioritizing Qualitative Taste Over Automated LLM Benchmarks
Automated benchmarks push AI models to produce average, middle-of-the-road results. If you replace standardized leaderboards with evaluations weighted by your own preferences, you can find the specific models that actually speed up your product development.
View Episode Notes →
Prioritizing Iteration Over Planning Using The Trashcan Method
Treat code as disposable to increase development speed. By skipping formal documentation and Jira in favor of rapid iteration, teams can prioritize immediate customer feedback over the slow, inaccurate process of upfront planning.
View Episode Notes →
Achieving Architectural Independence Through Open-Weight Coding Agents
High-performance coding agents are becoming a commodity, which allows teams to avoid proprietary vendor lock-in. By adopting open-weight models, you can maintain architectural independence and reduce costs without sacrificing output quality.
View Episode Notes →
Building Automated Security Pipelines Through Goal--Loop Harnesses
AI security succeeds when you build automated goal-loop harnesses rather than focusing on the model alone. By integrating agents into your existing developer tools and verification guardrails, your team can turn manual bug-finding into a high-throughput pipeline.
View Episode Notes →
Transitioning From Manual Prompting To Autonomous Agent Loops
Stop using manual, turn-based prompting. Move toward loop engineering to create autonomous agents that handle their own workflows. However, you must implement strict goal-based validation to avoid runaway token usage and systemic operational failure.
View Episode Notes →
Prioritizing Systematic Evaluation Over Manual Engineering Construction
Engineering leaders gain an advantage by moving away from manual coding and toward building rigorous evaluation pipelines. When teams treat automated testing as their primary product, they scale individual expertise and work faster than manual development cycles allow.
View Episode Notes →
Why Smarter AI Demands Narrower Use Cases
Claude Fable 5's superior intelligence creates hidden costs: its thoroughness drowns teams in dense, hard-to-parse output, while safety constraints silently stifle ambition. The real edge isn't raw power--it's using Fable 5 strategically, not dominantly.
View Episode Notes →
How AI Can Enforce Quality By Automating Values-Based Shopping
AI doesn't have to mean more mindless buying--Nicole Ruiz uses it to enforce a high-standards filter, making her purchase less but far better. Her system rewards durability, heritage, and integrity, turning curation into a quiet act of resistance against disposable culture.
View Episode Notes →
Ai Avatars Enable Real-Time Creative Iteration
AI-generated avatars aren’t about replacing you -- they’re about collapsing the time between idea and execution. The real advantage? Anyone can now iterate on video content in minutes, not months, unlocking a new creative flywheel.
View Episode Notes →
Non-Technical Founders Leverage AI for Rapid App Development
A beginner's mindset and precise prompting unlock AI-powered app creation, bypassing traditional technical barriers for rapid execution and validation.
View Episode Notes →
Opus 4.8's "Last 10%" Problem: Confidence Without Grounding
Opus 4.8 excels at rapid prototyping but struggles with the critical "last 10%" of development and grounding outputs in verifiable data, leading to potential hallucinations and a need for rigorous human oversight.
View Episode Notes →
AI Shifts From Assistant to Autonomous Agent via Goals
AI transitions from assistant to autonomous agent, tackling complex, multi-hour tasks independently. This shifts focus from constant prompting to achieving defined outcomes, freeing human cognitive resources.
View Episode Notes →
AI as Co-Creator: Beyond Automation to Augmenting Human Ingenuity
AI transforms human ingenuity by shifting focus from task execution to collaborative creation, unlocking novel solutions and personalized experiences beyond mere efficiency.
View Episode Notes →
Balancing Google's Rapid AI Velocity With Operational Instability
Google releases AI tools quickly, providing high performance but requiring significant operational effort. To succeed, treat these volatile tools as sandbox assets while using their native data access to build lasting competitive advantages.
View Episode Notes →
Replacing Markdown with HTML Artifacts for AI Planning
Markdown planning fails when AI projects scale. This results in ignored specifications and blind trust. By moving to interactive HTML artifacts, you turn planning into a modular workspace that keeps humans involved as active compute allocators.
View Episode Notes →
AI Transforms Engineering Through Spec-Driven Development and Collaboration
AI revolutionizes engineering by automating complex tasks, enabling spec-driven development, and transforming collaboration for increased joy and productivity.
View Episode Notes →
New Claude Features Enable Sophisticated Agentic Software Workflows
AI agents now self-grade work against rubrics, schedule tasks autonomously, and collaborate in teams, fundamentally changing how we build and deploy AI applications.
View Episode Notes →
Empowering Employees as AI Builders Drives Competitive Advantage
Empower every employee to build AI solutions, transforming your organization into an AI-native entity through an internal marketplace of needs and builders.
View Episode Notes →
Building Internal AI Tools for Design Fidelity and Efficiency
Bypass generic AI design outputs by embedding your design system into internal tools, transforming reviews and boosting product quality.
View Episode Notes →
Agentic Design and Vibe Coding Unlock Product Innovation
AI agents are becoming users, demanding API-first products over traditional UX. Empower non-technical teams to build directly, creating unique tools and a competitive edge.
View Episode Notes →
GPT-5.5's "Intelligence Tax" Justifies Solving Complex Tech Debt
GPT-5.5 unlocks autonomous problem-solving for complex tech debt and security challenges, justifying its "intelligence tax" by saving immense human engineering time on previously intractable issues.
View Episode Notes →
AI Design Tools: Speed vs. Iteration and Long-Term Integration
AI design tools offer rapid prototyping but hit critical limits. Discover the true cost of "instant" design and when human iteration still wins.
View Episode Notes →
Intercom Doubles Engineering Velocity Through AI-Driven Workflow Reimagining
AI is not just augmenting engineers; it's fundamentally re-architecting R&D workflows, doubling velocity by treating the engineering organization itself as a product.
View Episode Notes →
AI Orchestration: Building Persistent, Context-Aware Workflows
AI transforms knowledge work by orchestrating persistent, context-aware systems. Build your personal AI operating system to automate complex workflows and gain a significant advantage.
View Episode Notes →
AI-Powered Filtering of Overwhelming Digital Communication
Transform overwhelming digital communication into actionable insights with custom AI tools. Reclaim your workflow and gain a competitive edge by intelligently filtering and prioritizing information.
View Episode Notes →
Codebase as AI-Driven Customer Experience Engine
Your codebase is your most valuable asset when you know how to ask it the right questions. Discover how AI-powered IDEs transform customer-facing teams into proactive problem-solvers, delivering hyper-personalized solutions.
View Episode Notes →
AI-Driven "Anti-System" Reclaims Focus Via Observational Learning
Reclaim mental bandwidth by letting AI observe and adapt to your workflow, bypassing complex systems for a simpler, more focused approach to productivity.
View Episode Notes →
Stripe's Minions: AI Agents Reshape Engineering Workflows and Create Economic Actors
AI agents drastically lower the barrier to coding, freeing engineers for review and ideation. This shift enables non-engineers to contribute code and paves the way for autonomous AI agents as economic actors.
View Episode Notes →
AI Agents Automate Mundane Tasks, Reclaim Time for Strategy
Reclaim hours by automating mundane tasks with AI agents. Free your mental energy from operational friction to focus on strategic thinking and innovation.
View Episode Notes →
AI Agents Empower Non-Technical Builders to Ship Production Apps
Non-technical builders now ship production apps by orchestrating AI agents as a "picky customer," transforming software creation without traditional coding skills.
View Episode Notes →
AI Enables Fluid Design-Code Loop, Enhancing Planning and Craft
AI makes coding as cheap as designing, freeing teams to invest more in planning and craft for faster, higher-quality products.
View Episode Notes →
AI Brand Imagery: Visual Language Beats Simple Prompts
Unlock AI's creative potential by mastering visual language, not just prompts. Learn Gannon's iterative method to build consistent, on-brand imagery that commands trust and market recognition.
View Episode Notes →
Coinbase AI Adoption: Ditching Meetings for PR Velocity
AI transforms engineering by replacing meetings with massive PR velocity, freeing engineers from toil to accelerate product development and gain competitive advantage.
View Episode Notes →
Specialized AI Agents: Partitioning for Progressive Trust and Physical-Digital Integration
Partition specialized AI agents with dedicated hardware to prevent data leakage and enable personalized, human-like task management, unlocking new capabilities.
View Episode Notes →
Embrace Reality Early: Code-Based Prototyping for AI Products
Design AI products by embracing code-based prototyping and AI assistance to uncover flaws early, iterate rapidly, and build more user-centric outcomes.
View Episode Notes →
AI Empowers Individuals to Build Personalized Software Solutions
AI democratizes custom software, empowering individuals to build personalized tools rapidly and efficiently, bridging the "expert gap" and turning personal friction into a competitive advantage.
View Episode Notes →
AI's Dual Role: Generative Creation Versus Rigorous Code Refinement
Leverage AI's dual role in coding: use Opus for creative generation and Codex for rigorous refinement to build robust software faster and avoid common pitfalls.
View Episode Notes →
Building Custom AI Developer Tools for Enhanced Productivity
Build custom AI developer tools to unlock superior code quality and productivity. Move beyond generic solutions to craft intelligent environments that directly address your unique development challenges.
View Episode Notes →
AI-Powered v0 Integrates Git Workflows for Democratized Code Contribution
AI transforms code contribution, enabling marketers and designers to prompt production-ready features directly. Bypass traditional bottlenecks and accelerate innovation with AI-powered development environments.
View Episode Notes →
MCPs Unlock AI Potential Through App Integrations and Task Automation
Unlock AI's true potential by connecting it to your apps. This transforms tedious tasks into automated workflows, moving beyond simple prompts to intelligent agents that actively manage your digital life.
View Episode Notes →
Autonomous AI Assistants: Power, Peril, and Practical Integration
Autonomous AI assistants offer immense power but pose significant security risks. Understand the trade-offs before granting deep access to your data and systems.
View Episode Notes →
Proactive Context and Automation Elevate AI Coding Tool Productivity
Transform AI coding tools into powerful partners by pre-loading context with diagrams and automating quality checks, drastically accelerating development and improving reliability.
View Episode Notes →
Building Personalized AI Workflows for Research, Writing, and Task Management
Build personalized AI workflows to automate research, manage tasks, and enhance writing, transforming generic AI interactions into deeply integrated, efficient assistants.
View Episode Notes →
Codex Accelerates Development Through Detailed Planning and Parallel Workflows
Codex accelerates development by handling detailed implementation, enabling engineers to focus on strategic planning and architectural design. Automate code reviews and parallel workflows to boost productivity.
View Episode Notes →
AI Adoption Redefines Culture, Talent Sourcing, and Leadership
Discover your organization's unspoken culture and gain objective hiring insights by analyzing meeting transcripts and using AI interview evaluators. Unlock new efficiencies and find hidden talent.
View Episode Notes →
Executives Build AI Chiefs of Staff to Drive Organizational AI Adoption
Executives gain "brutal truth" feedback and automate tasks with a personal AI chief of staff, driving self-awareness and freeing up cognitive load for strategic work.
View Episode Notes →
Structured AI Adoption Drives Thought Leadership and Innovation
Unlock AI's power with a dual-learning approach and clear guidance. This strategy boosts adoption, mitigates risk, and positions your company as an innovative leader.
View Episode Notes →