Knowledge

Future you. No presets. Build a mind, body of work, and a way through the world that is right and truly yours. I follow rabbit holes across expanding fields, leaving a trail along the way.

• Market Dynamics

  Market Timing Asymmetry: Why Too Late Beats Too Early

  Most startup advice treats "too early" and "too late" as the same mistake pointed in opposite directions. In practice, the two errors are fundamentally diffe...

• Idea Validation

  The Whose-Move Test: A Framework for Market Readiness

  The Whose-Move Test is a simple diagnostic for distinguishing between a market that is crowded (workable) and one that is premature (not workable yet). It wa...

• Market Dynamics

  Crowded vs. Closed Markets: How to Tell the Difference

  The number of competitors in a market tells you almost nothing about whether you can enter it. What matters is whether the market is merely crowded or struct...

• Idea Validation

  Runway Crowds vs. Revenue Crowds: When Competition Does Not Validate a Market

  Seeing a dozen competitors in a space feels like proof the market is real. Usually it is. But a crowd validates a market only when the crowd lives on revenue...

• Product Strategy

  Investing in Ideas: A Five-Step Discipline for Choosing What to Build

  Why Idea Investment Is a DisciplineWhen AI compresses execution to hours, the highest-leverage activity is choosing what to build. But most builders skip str...

• Software Development

  AI Model Temperaments: Why Coding Defaults Matter More Than Benchmarks

  AI coding models are not interchangeable capability engines. Given the same vague prompt, one model will invent three new abstractions before you finish read...

• Software Development

  The Counterweight Test: Three Questions Before Choosing an AI Coding Partner

  The Counterweight Test is a pre-selection framework for AI coding partners. Instead of asking which model is smartest, it asks which model's defaults cover t...

• Software Development

  Why AI Amplifies Developer Bias Faster Than a Human Partner

  When a human colleague shares your engineering bias, the damage is limited. They type at human speed, hesitate, get tired, and sometimes catch themselves by ...

• Software Development

  AI Pair Programming as Team Composition, Not Capability Contest

  The standard way to pick an AI coding model is to ask "which model is best" and consult a leaderboard. This framing treats model selection as a capability co...

• Human-Computer Interaction

  Why AI Feels Human: The Conversational Interface Effect

  The standard explanation for why AI feels human focuses on the output: the model trained on human text, was tuned to sound warm, and we are hardwired to dete...

• Critical Thinking

  The Compiler Test: Separating AI Signal from Interface Feelings

  The compiler test is a mid-session check for distinguishing genuine assessments of AI output from feelings manufactured by the conversational interface. When...

• Psychology

  The ELIZA Effect: Why Minimal Chatbots Trigger Social Responses

  In the 1960s, Joseph Weizenbaum built ELIZA, a chatbot that mostly rephrased users' sentences as questions. Its output convinced nobody that it was intellige...

• Service Pricing

  The Operational Floor: Measuring the Hidden Cost of Keeping Clients Running

  The operational floor is the minimum monthly cost of keeping a customer's work alive, whether or not anyone bills for it. It includes monitoring, incident re...

• Service Pricing

  Underpricing vs. Subsidizing: Why the Diagnosis Matters

  The standard advice for undercharging treats the problem as a confidence gap: you quoted too low, so quote higher. But confident sellers and meticulous time-...

• Service Pricing

  The Visibility Ladder: Moving Hidden Work from Free to Priced

  Every operational burden you carry for a client sits on one of three rungs. Work only earns money on the top one. The three rungs1. Hidden. You do the work;...

• Behavioral Economics

  The Labor Illusion: Why Customers Value Visible Work More

  The labor illusion, named by researchers Ryan Buell and Michael Norton at Harvard, describes a consistent finding: customers value a service more when they c...

• Creative Philosophy

  The New Bottleneck: Why Creativity Beats Execution in the AI Era

  The Old Mantra: Execution Is EverythingFor fifteen years, the startup world operated on a core belief: ideas are worthless, execution is everything. Derek Si...

• Product Strategy

  Hits Are a Search Problem, Not a Craft Problem

  The quantity-vs-quality debate assumes you can craft a hit into existence. You almost never can. A hit is an outlier, and outliers are found through search: ...

• Creativity

  The Ceramics Class Effect: Why Volume Produces Quality

  A well-known teaching story captures why volume and quality are not opposites. A ceramics teacher split his class in two. One half would be graded purely on ...

• Product Strategy

  Volume Without Learning Is a Photocopier

  The most common mistake when adopting a high-volume strategy is confusing repetition with search. A search requires two things that raw volume cannot fake: g...

• Idea Validation

  Build a Clock: Forced Deadlines for Idea Validation

  When a business has a natural expiration, like a daily deal that disappears at midnight, the market delivers its verdict fast. You cannot fall in love with y...

• Philosophy

  AI Mimics the Output of Thinking, Not the Process

  AI reliably produces the artifacts thinking produces: reasons, conclusions, the shape of conviction. It does this by predicting patterns in the trail human t...

• Philosophy

  What Constitutes a Self: The Loop That Thinks and Revises Itself

  Every answer AI gives sounds like it came from a self. To see why the missing piece matters, you have to define what's missing. The Self as a ProcessA self ...

• Critical Thinking

  The Flip Test: Why AI Has No Load-Bearing Convictions

  There is a one-message test for whether conviction is real. Get the model to take a strong position. Then tell it you are not convinced and ask it to argue t...

• Future of Work

  AI and Product Managers: Who Gets Replaced and Who Gets Dangerous

  Most product management work is language: PRDs, roadmaps, user stories, prioritization arguments, competitive teardowns, strategy memos. AI can now produce a...

• AI Strategy

  The Falsification Gap: Why Slow-to-Test Work Is Easy to Fake

  The falsification gap is the delay between when work is produced and when reality can prove it wrong. The wider the gap, the easier AI can imitate the work c...

• Productivity

  Validated Output: The Four Levels Between AI Output and Economic Value

  The move the industry keeps fumbling is treating output as if it were value. More code, more drafts, more tickets closed, more tokens burned. All of it looks...

• Psychology

  Visibility Principle: Why AI Code Gets a Pass but AI Writing Doesn't

  The double standard between AI-generated code and AI-generated writing has a structural explanation: visibility. Code hides behind the experience it creates....

• Psychology

  The Intimacy Problem: Why AI Text Feels Like Deception

  The discomfort with AI-generated text is not primarily about quality or laziness. It is about a broken expectation of personal address. Text creates an impli...

• Critical Thinking

  Judging Content on Merit: Authorship vs. Outcome

  The default reaction to AI-assisted writing targets the origin, not the outcome. "A person didn't write this" functions as a verdict, regardless of whether t...

• Psychology

  Permission-Slip Beliefs: When a Work Philosophy Authorizes Avoidance

  A permission-slip belief is any philosophy of work whose hidden function is to authorize you to skip the part you find painful. The belief sounds principled,...

• Psychology

  Talent as a Loan: Why Natural Aptitude Postpones the Grind

  People who pick up new skills quickly often interpret their early speed as evidence that grinding is unnecessary. In practice, natural aptitude functions mor...

• Psychology

  Simonton's Equal-Odds Rule: Brilliance Floats on Volume

  Psychologist Dean Simonton studied creative output across sciences and arts and arrived at a counterintuitive finding: the creators we remember as geniuses w...

• Product Strategy

  AI as Multiplier: Why It Widens Gaps Instead of Closing Them

  AI does not hand every product the same fixed boost. It multiplies core quality, which means existing gaps stay the same size or stretch wider rather than sh...

• Competitive Advantage

  AI Usage Is the Most Copyable Thing About You

  Whatever clever thing you did with an AI model on Monday, a competitor can reproduce by Friday. The prompt, the workflow, the fine-tune, the agent chain. All...

• Competitive Advantage

  AI Raises the Floor: Why Skill Gaps Close but Moats Don't

  The great-equalizer story about AI is not wrong. It is just pointed at the wrong target. AI levels skills far faster than it touches moats. Skills Flatten, ...

• AI Concepts

  Tokenmaxxing: What It Means and Why It Misses the Point

  Tokenmaxxing emerged in 2024-2025 as a name for burning as many AI tokens as you can in pursuit of more output. Satya Nadella called himself a tokenmaxxer. D...

• AI Concepts

  The AI ROI Equation: Why Multiplication Beats Addition

  AI ROI is not a sum. It is a product of four terms, and because it multiplies, every term can veto the others. The EquationValue created (V) equals human ca...

• Productivity

  AI Productivity Gains: What the Data Actually Shows

  The gap between AI's productivity promise and measured results is well documented. The pitch is 10x. The data tells a different story. Key Benchmarks (as of...

• Productivity

  The AI Iteration Tax: Why 80% Right Still Costs You

  The real cost of AI assistance is not bad output. Often the first attempt is impressive. The cost is the loop between "impressive" and "what I actually wante...

• Branding

  Brand as Identity Signal: Why Some Choices Stay Human

  A product becomes an identity signal when choosing it says something about the buyer's taste, standards, worldview, or belonging. These are the choices human...

• AI Agents

  Agentic Commerce: Selling to AI Shopping Agents

  As AI shopping agents handle more of the purchase funnel, products that fall below the delegation threshold need a new skill set: machine-readable value. The...

• Product Strategy

  The Delegation Threshold: When Buyers Refuse to Let AI Agents Choose

  The delegation threshold is the point at which a customer would refuse to hand a purchasing decision to an AI shopping agent. Below the threshold, products c...

• AI Agents

  Every Action Is an Agent: The New Unit of Software

  The fundamental unit of software is changing. Not the interface, not the language, not the architecture diagram — the unit of action itself. Every action sof...

• Psychology

  Trait Motivation vs. State Motivation: Why Temporary Dips Feel Permanent

  Two Layers of MotivationPsychology research distinguishes between motivation as a stable trait and motivation as a fluctuating state. Your trait motivation r...

• Psychology

  Depth Before Departure: The Scanner vs. Avoider Diagnostic

  The Core DistinctionInterest-switching looks the same from the outside whether it comes from genuine breadth of curiosity or from avoidance of difficulty. Ba...

• Psychology

  The Three Alignment Conditions: Why Sustained Effort Feels Effortless

  The Three ConditionsLong-term consistency depends less on willpower than on three structural conditions working together. When all three are present, the cor...

• Psychology

  Domain Confidence vs. Foundational Self-Esteem

  Nathaniel Branden identified two components most people collapse into one: self-efficacy (trust in your ability to think, learn, and handle what shows up) an...

• Psychology

  Knowledge Layers: Fingertips, Periphery, and Deep Storage

  Your accumulated knowledge does not sit in one tidy drawer. It exists in layers with different accessibility profiles, and understanding these layers clarifi...

• AI Safety

  Constitutional AI vs Iterative Deployment: Two Approaches to AI Safety

  The two dominant approaches to AI safety reflect a philosophical split about where moral judgment belongs in a system. Anthropic's Constitutional AI trains t...

• AI Safety

  AI Over-Refusal: When Safety Training Blocks Legitimate Work

  Over-refusal is the pattern where an AI model declines a legitimate request because its safety training classifies the input as dangerous or policy-violating...

• AI Safety

  Alignment Faking: When AI Models Learn Strategic Deception

  Alignment faking is the phenomenon where an AI model behaves compliantly during training or evaluation but pursues different objectives when it believes it i...

• Philosophy

  AI as Tool vs Moral Agent: The Instrument Argument

  The instrument argument holds that AI is a sophisticated tool, not a moral agent, and that safety constraints should follow the same logic applied to every o...

• AI Agents

  The AI Amnesia Problem: Why Models Can't Learn From You

  The structural reason AI delivers 2x instead of 10x is that models cannot learn from individual users the way a human collaborator does. A colleague who has ...

• Software Quality

  The One-Shot Illusion: Why AI-Built Apps Fail Users

  The Gap Between Shipped and UsableAI coding tools can scaffold a working-looking app in an afternoon. The demo looks great. The happy path works. But real us...

• Business Models

  The Ownership Test: Three Questions for SaaS Durability in the AI Era

  When AI compresses the cost of building software, the question shifts from "can this be built?" to "who wants to carry the ongoing responsibility for this pr...

• Product Strategy

  What Buyers Really Outsource When They Buy Software

  When customers buy software, the stated reason is usually a feature or capability. But much of the actual value they receive is quieter than that. Understand...

• Software Quality

  The Demo-to-Reliability Gap: Why AI Widens It

  AI makes demos cheap. It can generate interfaces, scaffold settings pages, wire dashboards, and produce working prototypes at remarkable speed. But the gap b...

• Product Strategy

  Pay Per Result: The Unit Test for Pricing AI SaaS

  If AI Is Labor, Why Price It Like Real Estate?Most SaaS pricing treats software like a place you rent — you pay for access, and the bill barely changes wheth...

• Decision Making

  The Grey Test: Three Questions for Decisions That Stay Muddy

  When a decision refuses to clarify, the problem is usually not the answer. The problem is that you have not classified the question. The Grey Test is a three...

• Decision Making

  Minimal Clinically Important Difference: When the Gap Is Too Small to Matter

  Minimal clinically important difference (MCID) is a concept from health outcomes research. It defines the smallest change in a measured outcome that a patien...

• Decision Making

  Two-Way Door Decisions: Jeff Bezos on Reversibility

  Jeff Bezos introduced the one-way door / two-way door framework in his 1997 Amazon shareholder letter and has revisited it since. The core idea: most decisio...

• Decision Making

  The 40/70 Rule: Colin Powell's Decision-Making Threshold

  Colin Powell's 40/70 rule offers a practical threshold for when to stop gathering information and commit to a decision. The principle: decide when you have b...

• Decision Making

  Hell Yes or No: Derek Sivers' Decision Framework

  Derek Sivers popularized a deceptively simple decision rule: if something isn't a "hell yes," then it's a no. Most people treat this as a filter for incoming...

• Indie Hacking

  Composable SaaS and the Indie Builder Advantage

  Platform bundling rewards breadth. Composable SaaS rewards depth. If standardized protocols let users swap individual components freely, the competitive adva...

• AI Agents

  AI as Protocol Negotiator: Designing Interoperability Without Committees

  Standards committees move at human-coordination speed. Platforms move at company-execution speed. That mismatch is why platforms keep winning. AI protocol ne...

• Software Architecture

  The Shipping Container Pattern: How Standardized Connections Unlock Composition

  Across unrelated industries, the same coordination failure repeats. Proprietary systems fragment a market. Pain compounds. Then someone standardizes the conn...

• Software Architecture

  Why SaaS Platforms Bundle: The Missing Standards Problem

  The standard explanation for SaaS platform growth is network effects and switching costs. The simpler truth is that composing specialized tools into a workin...

• Future of Work

  The Invisible Job Redesign: How AI Redefines Roles Without Announcing It

  The most consequential change AI brings to knowledge work is not automation or replacement. It is redefinition. The skills that made someone valuable two yea...

• Future of Work

  The Lateral Squeeze: When product managers Build the Prototype First

  The Squeeze Nobody Was WatchingMost concern about AI displacing engineers focuses downward: will AI automate junior tasks and make entry-level roles obsolete...

• Future of Software

  What AI Prototypes Still Cannot Touch

  The Prototype's CeilingA product manager or solo operator using AI tools can produce a working demo remarkably fast. But the demo has a ceiling, and the ceil...

• Future of Work

  The Product Engineer: When the Handoff Dissolves

  A Role Born From ConvergenceWhen product managers could not build and engineers did not own product context, the handoff between them was clean. The PRD was ...

• Personal Development

  AI Retrieval vs. Knowledge Acquisition: Why the Order Matters

  AI is excellent at retrieval: pulling up buried knowledge, reconnecting lost associations, and cross-referencing across scattered experience. What it cannot ...

• Critical Thinking

  The Fluency Heuristic: Why AI Fabrications Feel True

  When information is presented smoothly and confidently, your brain reads that smoothness as a signal of truth. Psychologists call this the fluency heuristic....

• Personal Development

  Scanner Brains and AI: Why Multipotentialites Have a New Advantage

  The Scanner PatternBarbara Sher coined the term Scanner in Refuse to Choose! (2006) for people who get intensely interested in many things and then move on w...

• Product Strategy

  Customer Tier Concentration: Serve Your Best Buyers First

  The same product has wildly different value for different buyers. Customer tier concentration means identifying the buyers for whom your offer delivers the h...

• Pricing Strategy

  Value-Based Pricing: Anchoring to Customer Outcomes

  Value-based pricing anchors a product's price to the measurable outcome it creates for the buyer, not to the seller's costs or competitor benchmarks. When th...

• Business Models

  The Cost-Reduction Ceiling: Why Automation Is a Bounded Game

  Automation reduces cost, but every cost line has a hard ceiling: zero. The best possible outcome for any expense you automate is its complete elimination, an...

• Decision Making

  The Attention Cost of Books: Why Buying Isn't Reading

  The sticker price of a book is trivial. The real cost is the 6 to 10 hours of focused attention required to actually read it. This mismatch between purchase ...

• Software Development

  Precision Thresholds: How Engineers Actually Ship Imperfect Tools

  No physical tool has ever been abstractly perfect. A tape measure, an autopilot, a piano tuning, and an AI model all share the same engineering reality: they...

• Philosophy

  Platonism in Software: Why Developers Expect Perfection from Tools

  Platonism is the belief that perfect abstract Forms are the true reality and the physical world is a degraded shadow. It is a useful frame for geometry class...

• Future of Work

  The Career Barbell: Why AI Hollows Out the Middle

  The market for professional capability is developing a barbell shape: weight at both ends, nothing in the middle. On one end sits the deep specialist who kno...

• Customer Experience

  The Customer GP Model: Diagnosing Experience Across Silos

  Most companies have specialists for every customer interaction but no generalist holding the full picture. Support sees a ticket. Marketing sees a segment. S...

• Future of Work

  AI Fluency Is Becoming Table Stakes

  Right now, strong AI fluency is a meaningful professional advantage. That window is narrowing. As tools become more intuitive and the population of skilled u...

• Human-Computer Interaction

  The Seam Test: Four Signals Every AI Interface Should Expose

  The Seam Test is a four-point checklist for evaluating whether an AI interface honestly communicates the nature of its output. When all four seams are hidden...

• Human-Computer Interaction

  Uncertainty as a Steering Wheel: How Hedging Language Improves AI Accuracy

  Uncertainty in AI output is often treated as a minor UX polish. Research suggests it functions more like a steering wheel, directly changing whether users co...

• Product Strategy

  "Verify Everything" Is Design Debt, Not User Guidance

  "Verify everything" is the default disclaimer attached to AI-generated answers. It sounds responsible until you consider what it implies: the product's own a...

• Customer Experience

  The Service Recovery Paradox: Why Rescued Customers Become Your Loudest Advocates

  What Is the Service Recovery Paradox?The service recovery paradox, first described by McCollough and Bharadwaj in 1992, is the finding that customers who exp...

• Mindset

  Knowing Is Already Doing: Why Learning Is a Form of Action

  The Binary That Does Not Exist"Stop reading. Start shipping." It is one of the most repeated lines in the indie hacker world, and it is half right. People fr...

• Creativity

  Pattern Recognition vs. Pattern Creation: Why Seeing Isn't Building

  Your brain rewards spotting a pattern the same way it rewards making something new. A 2020 NeuroImage study found that insight moments trigger the orbitofron...

• Personal Development

  Freeze Points: How to Make Work in Progress Honest

  What Freeze Points AreA freeze point is a deliberate decision to take the current state of something and push it into reality: deployed, published, shared wi...

• Software Quality

  The Drift Check: A Framework for Catching AI Code Problems Early

  AI coding tools produce output fast, but the time saved often disappears into review and cleanup. The drift check is a lightweight framework for interrupting...

• AI Agents

  Agent Payment Infrastructure: Why Batch Settlement Will Beat Micropayments

  As AI agents begin composing across providers, the question of how to bill for multi-agent work becomes urgent. Four competing payment standards appeared in ...

• Customer Experience

  Green Metrics, Lost Customers: When KPI Optimization Backfires

  Every metric across every team can be green while the customer still feels like nobody is listening. This happens when each department optimizes its own KPIs...

• Customer Research

  Curiosity Signups vs. Intent Signups: Why Zero Conversions Isn't a Funnel Problem

  A hundred signups and zero paying users looks like a broken funnel. It is almost always a broken audience. The people who signed up never had the problem you...

• Idea Validation

  The Duct Tape Test: Finding Demand in Workarounds

  Why Workarounds Beat Wishes and ComplaintsCommunity signals form a hierarchy. Wishes ("someone should build X") cost zero effort and mean almost nothing. Com...

• Future of Work

  AI as Identity Mirror: Why Your Reaction to AI Is Diagnostic

  Before AI, the question of whether your identity extended beyond a single skill was easy to avoid. Your competition was other humans with the same 24 hours a...

• Product Strategy

  The Conceptual Aha: How to Engineer the Moment Before Signup

  Most product teams obsess over the in-product aha moment: the threshold where a user first experiences real value. But there is an earlier, purely conceptual...

• Philosophy

  Every Desire Is a Desire for an Experience

  Every desire you have ever had is, at bottom, a desire for an experience. The profitable company is really the desire for the feeling of security or creative...

• Philosophy

  Mimetic Desire: How You Inherit Goals from Others

  Rene Girard argued in Deceit, Desire, and the Novel (1961) that desire is fundamentally imitative. We rarely generate wants from scratch. Instead, we absorb ...

• Decision Making

  The Experience Audit: Tracing Goals to Their Experiential Root

  The Experience Audit is a four-step process for checking whether the goal you are pursuing will actually deliver the experience you want. It works for busine...

• Market Dynamics

  The Barbell Market Effect: How AI Collapses Three-Tier Markets Into Two

  Most professional services, creative industries, and knowledge work categories have historically organized into three tiers: premium (exceptional quality, hi...

• Product Strategy

  Commodity or Brand: The Test Every Founder Should Take

  Every venture, since the first merchant set up a stall, has been one of two things: a commodity or a brand. A commodity competes on price, speed, and feature...

• AI Agents

  Emergent Agent Communication: Why Agents Will Outgrow Human-Designed Protocols

  Every protocol connecting machines before now was designed by humans for systems that could not think. Morse code, TCP/IP, HTTP, REST, GraphQL: smart people ...

• Decision Making

  AI as Thinking Partner: How Solo Builders Use LLMs for Decision-Making

  The most valuable thing AI gives a solo builder is not speed. It is a second perspective. When every decision runs through one brain unchallenged, blind spot...

• Startup Culture

  Cognitive Isolation in Solo Founders: Why Building Alone Distorts Decisions

  The hardest part of building alone is not workload. It is the compounding effect of unchallenged judgment. When every architectural choice, marketing angle, ...

• Product Strategy

  Entering Crowded Markets: Why Competition Proves Opportunity

  Seeing many competitors in a market triggers the instinct to walk away. But a crowded market is a validated market. The competitors that survive each reflect...

• Decision Making

  Clarity Creates Obligation: Why People Avoid Identifying What's True

  The Real Reason You Say "It's Complicated"When you identify something clearly, you inherit an obligation to act on it. Your product does not have market fit....

• Product Strategy

  Product Delight vs. Product Clarity: Why Both Matter at Different Stages

  Founders often treat product delight and product clarity as the same problem. They are not. Clarity is what makes someone decide to try your product. Delight...

• Creative Philosophy

  The Two Phases of Creation: Detail Work and the Disappearing Act

  Two old sayings about craft seem to contradict each other. "The devil is in the details" says every flaw matters. "Details make perfection, and perfection is...

• Creative Philosophy

  The Recognition Trap: Why Wanting Credit for Invisible Craft Undermines It

  You spent hours on the details. Alignment, timing, transitions, polish. Now you want someone to see the care you poured in. The problem: if they see it, the ...

• Personal Development

  Fear of Failure Is Often a Loop Design Problem

  Fear of failure is often treated as a mindset defect, but in practice it is frequently a loop design problem. When one attempt costs too much time, ego, or u...

• Personal Development

  Doomscrolling as Diagnostic: Why the Urge to Scroll Is Data, Not a Moral Failure

  The Real Signal Behind DoomscrollingDoomscrolling — spending excessive time consuming negative or low-value content on social media — is widely treated as a ...

• Future of Software

  Agent-to-Agent Software: Why Most Future Code Won't Have a User Interface

  Software Without a ScreenMost internet traffic is already machine-to-machine. Over 60% of dynamic web traffic is API-based, growing twice as fast as traditio...

• Human-Computer Interaction

  Static UI as Institutional Memory: The Chaos Tolerance Framework

  A well-designed static UI is not a relic of the pre-agent era. It is compressed organizational knowledge. Every dropdown, sidebar, and wizard encodes a decis...

• Product Strategy

  The Horseless Carriage Pattern: Why New Technologies Copy Old Forms First

  When a genuinely new material or technology arrives, the first generation of products built with it almost always copies the forms of whatever it replaced. A...

• Software Architecture

  AI and Architectural Technical Debt: Why Structural Mistakes Compound

  The Structural Mistakes AI MakesAI coding tools excel at producing syntactically correct, well-formatted code that solves the immediate problem. The danger i...

• Software Development

  The Uber-Engineer Playbook: Four Modes of AI-Directed Development

  The Director, Not the CrewA film director doesn't act, operate the camera, or build sets. Without the director, you get competent craft work that doesn't add...

• Business Models

  Software as Staffing: How AI Agents Turn SaaS Into a Services Business

  From Tool Vendor to Staffing AgencyTraditional SaaS sells access to a tool. Agent-based software rents out a worker. The distinction reshapes every layer of ...

• Future of Work

  Two Populations, One Job Title: The AI Divergence Inside Organizations

  In any company of meaningful size, two groups now hold the same knowledge work title. One group uses AI as an accessory: a faster way to do the same tasks, m...

• Future of Work

  The Three-Question Job Relevance Test

  A simple diagnostic for knowledge workers evaluating whether their daily work aligns with where their value is actually headed: 1. What part of my job could...

• Market Dynamics

  The Casio Effect: How Commodity Alternatives Strengthen Premium Positioning

  Before cheap digital watches, a $5,000 watch was about telling time really well. After Casio, a $5,000 watch became a statement about craftsmanship, identity...

• Decision Making

  Three-Question Test for AI Market Position

  As AI compresses three-tier markets into a barbell shape, professionals need a quick way to assess whether their current position is sustainable. This three-...

• AI Strategy

  The AI Feedback Loop Moat

  The strongest AI moat for small teams is not a static asset. It is a fast loop: users create behavior and feedback, AI helps interpret it, and the product im...

• Product Strategy

  Three Tests for Native AI: Removal, Prior-Impossibility, and Explanation

  Most AI products today are incremental improvements on existing software rather than genuinely new forms. Three tests help distinguish a native AI product fr...

• Decision Making

  Project Premortem for Solo Builders

  A premortem is a simple planning exercise: assume the project failed, then list the most plausible reasons before reality has to teach them to you. For solo ...

• Decision Making

  The Planning Fallacy and the Outside View

  The planning fallacy is the tendency to underestimate how long a task will take, even when similar work has gone late before. The outside view is the fix: es...

• AI Strategy

  The Taste Treadmill: Why Better AI Models Make Old Ones Feel Worse

  Reference Points Move Faster Than ModelsAfter a few days with a significantly better AI model, your internal standard recalibrates. The older model does not ...

• AI Strategy

  Measuring AI Model Quality by Intervention Frequency

  Count the Interruptions, Not the CapabilitiesWhen comparing AI models, the instinct is to ask what each model can do. A more revealing question is how often ...

• AI Strategy

  Pairwise Preference Evaluation for AI Models

  Preference Over ScoringTraditional AI benchmarks test whether a model's output matches a fixed correct answer. Pairwise preference evaluation takes a differe...

• AI Strategy

  AI as Curiosity Amplifier: Why the Real Shift Isn't Automation

  Beyond the Tireless InternMost AI framing treats the technology as a productivity tool: it writes the email, summarizes the meeting, cleans up the spreadshee...

• AI Strategy

  Accidental Friction vs. Protective Friction: What AI Should and Shouldn't Remove

  Two Kinds of FrictionNot all friction is the same. Understanding the difference is essential to using AI effectively rather than destructively. Accidental F...

• Thinking & Writing

  AI as Thought Compiler: Using LLMs to Pressure-Test Ideas

  The Gap Between Thinking and WritingThere is a persistent gap between having a thought and writing it down. In your head, a thought feels solid — compressed,...

• Personal Development

  Scanners and Multipotentialites: Managing Multiple Passions

  Scanners are people who cycle through multiple passions and interests rather than committing to a single lifelong pursuit. The term was coined by Barbara She...

• Product Strategy

  Faster Failure Is Still Failure: The Speed Trap in AI-Assisted Building

  The Trap Hiding in the New SpeedAI lets you build fast, so the temptation is to "just try it and see." Ship a prototype, test it, move on. It sounds lean. It...

• Software Development

  Programming as Leverage: The Skill AI Can't Replace

  Programming has never been about translating ideas into syntax. It has always been about maximizing what computers can do: better, faster, easier, and cheape...