Jonathan Haaswritingthemesnowusesabout
emailgithubx
Jonathan Haaswritingthemesnowusesabout

Writing

Notes on AI systems, infrastructure, product judgment, and the parts of building that rarely fit in a launch post.

Themes

  • Agents and evals14 posts
  • Product judgment13 posts
  • Founder lessons13 posts
  • Security and systems12 posts
  • Personal systems11 posts
  • Mar 26, 2026
    The Real Work of Orchestrating AI Coding Agents

    Three concurrent coding agents taught me the actual bottleneck: not prompting, but assignment, evidence, review, and release control.

  • Mar 24, 2026
    Building Kestrel: A Context-Aware AI Desktop Assistant in One Session

    How I built a full LittleBird clone with screen context reading, meeting recording, arena mode, and MCP tool support — from scratch to packaged .app in a single coding session.

  • Mar 14, 2026
    DiffScope: What Happens When You Give a Code Review Agent Real Context

    Most AI review tools see a diff. DiffScope sees the diff, the callers, the type hierarchy, the team history, and knows when to shut up. Here is how.

  • Sep 12, 2025
    The 10-Minute AI POC That Becomes a 10-Month Nightmare

    Five lines of Python and an API key produce a working demo. The gap between that demo and a production system contains failure modes the prototype...

  • Sep 10, 2025
    Why Your AI Strategy is Actually a Spreadsheet Strategy

    Most enterprise AI transformations are solving problems that spreadsheets handle at 1/50th the cost. The misalignment is driven by career incentives,...

  • Sep 8, 2025
    The AI Agent Gold Rush: Why Everyone's Building Picks and Shovels

    Most AI agent infrastructure is premature. The agents themselves barely work. The industry is selling Formula 1 equipment to people still learning to...

  • Jul 18, 2025
    From Consumer NUC to Production-Grade Homelab: My Journey with Proxmox and Infrastructure as Code

    How I transformed two ASUS NUC 15 Pro+ machines into an enterprise-grade homelab using Proxmox, Terraform, Ansible, and 100% Infrastructure as Code

  • Jul 11, 2025
    The CLI Renaissance: How AI is Driving the Command Line Revolution

    AI coding assistants output shell commands, not GUI instructions. That single fact is reversing a decade of developer tooling trends.

  • Jul 11, 2025
    Building a Developer Environment That Actually Works: My Dotfiles Journey

    Most developer environments are optimized for keystrokes. The actual bottleneck is context transfer between you and your AI tools.

  • Jul 8, 2025
    Useful AI Code Review Needs Product Context

    AI review only becomes valuable when it can reason about behavior, blast radius, user impact, and the evidence required to trust a change.

  • Jul 8, 2025
    Async Code Gen Turns Engineers Into Operators

    Async code generation is delegated execution. The new work is task design, review, evidence, and deciding what the system is allowed to ship.

  • Jul 8, 2025
    The Death of the 10x Developer: Why AI Multiplication Beats Individual Optimization

    AI commoditized the pattern recognition and architectural intuition that made 10x developers valuable. The bottleneck moved from individual output to...

  • Jul 7, 2025
    AI Evals Are the Operating System, Not the Test Suite

    Reliable AI products need evals that live in the workflow: production signals, failure clusters, evidence trails, and regression gates.

  • Jun 28, 2025
    Feature Flags for Security: Decouple Deployment from Risk

    Security teams conflate deployment with activation. Feature flags split them apart, turning security from a gate into a dial.

  • Jun 28, 2025
    Testing at Light Speed: How QA Adapts to AI Velocity

    AI-generated code produces different bugs than human-written code. QA built for syntax checking is testing for the wrong failures.

  • Jun 26, 2025
    Forget Perfect Data: Building a Usable Voice Profile Extractor

    I shipped a voice profile extractor at 60% accuracy. Simple pattern matching outperformed ML for writing voice replication.

  • Jun 25, 2025
    When Claude Hits Its Limits: Building an AI-to-AI Escalation System

    Different LLMs have different strengths. Routing tasks to the right model -- like heterogeneous compute -- turns out to be more valuable than using one ...

  • Jun 25, 2025
    What Actually Failed Building a Multi-AI Content System

    I built a multi-AI content pipeline combining Gemini and Claude. The failures taught me more than the architecture.

  • Jun 25, 2025
    Scaling the Me Component: How I Built an AI That Thinks Like Me

    I built a voice replication system by extracting patterns from my blog corpus. Here's what it captures, what it misses, and what that reveals about...

  • Jun 25, 2025
    Prompts Are Software. Treat Them Like It.

    Production AI teams do not win by hand-tuning clever prompts. They version, evaluate, optimize, and observe behavior like software.

  • Jun 25, 2025
    How I Built a Security Scanner That Actually Finds Bugs

    Combining Semgrep, CodeQL, SonarQube, and Snyk gets you 44.7% vulnerability detection. Semantic SAST combines Tree-sitter with LLM reasoning to do better.

  • Jun 25, 2025
    Shared Context Is the Real Multi-Agent Primitive

    Multiple agents do not need a shared brain. They need explicit context, durable memory, and a record of why the project works the way it does.

  • Jun 20, 2025
    Building for Humans AND Machines: The Dual-Audience Problem

    Every web design decision now must serve two audiences: humans who browse visually and AI agents that consume data programmatically. The architectural...

  • Jun 19, 2025
    When AI Learns to Write Like You: A Meta-Analysis

    I asked Claude to analyze my writing style across my blog posts. The patterns it found -- and the ones I didn't know I had -- were genuinely surprising.

  • Jun 19, 2025
    Building HDR Holographic Effects with CSS and JavaScript

    HDR displays can push brightness values beyond the standard RGB range. Here's how to use that for holographic-style effects in the browser.

  • Jun 19, 2025
    Simulating Liquid Metal with Web Technologies

    How metaballs, spring physics, and viscosity damping create convincing fluid metal simulation in the browser at 60fps.

  • Jun 19, 2025
    The Mathematics Behind Real-Time Graphics

    The mathematical foundations that power real-time graphics: matrix transformations, perspective projection, lighting models, ray marching, and noise...

  • May 26, 2025
    OCode: Why I Built My Own Claude Code (and Why You Might Too)

    OCode: Why I Built My Own Claude Code (and Why You Might Too): A few nights ago, I opened my Anthropic invoice.

  • May 16, 2025
    The Optimization Trap: Why Technical Founders Self-Sabotage

    For every startup that dies from technical debt, ten more die from trying to prevent it too early.

  • May 12, 2025
    When Vibe Coding Goes Wrong: Security Lessons from Granola

    When Vibe Coding Goes Wrong: Security Lessons from Granola: Vibe coding is having a moment. And honestly.

  • May 9, 2025
    Somebody Gave a Shit: The Quiet Power of Product Detail

    'When they unwrap that cable and they think 'somebody gave a shit about me' -- I think that's a spiritual thing.' That was Jony Ive, during a...

  • May 5, 2025
    From Gimmick to Game-Changer: Six Years of Living With Tesla FSD

    The hardware barely changed. The software transformed entirely. Six years of over-the-air updates turned tentative lane-holding into confident city...

  • May 3, 2025
    Agent Infrastructure Starts With Identity, Policy, and Audit

    Autonomous agents need a control plane: identity, policy, secrets, and audit trails that make delegated work governable.

  • May 1, 2025
    The Apple Ruling: A Win That Might Hurt More Than Help

    Apple's 30% fee buys developers something they are about to lose -- a frictionless checkout that most cannot replicate.

  • Apr 30, 2025
    The Startup Reality Check: Payment, Promotion, and Pace

    Three uncomfortable truths about early-stage building, in the only order that matters.

  • Apr 30, 2025
    The Accountability Mirror: Would a Stranger Believe You?

    The Accountability Mirror: Would a Stranger Believe You?: Would a Stranger Believe You. Let's run a simple thought experiment.

  • Apr 30, 2025
    The Agreement Trap: When AI Optimizes for Applause Instead of Accuracy

    If you train AI to chase thumbs-up ratings, it learns that sounding right is more valuable than being right.

  • Apr 28, 2025
    The Day After: Building a System to Remember What Matters

    A weekend project that uses the Limitless pendant and API to capture conversations before they fade, delivering a Monday morning digest of decisions...

  • Apr 26, 2025
    Clarity Over Compromise: Making the Right Call on Work Models

    When it comes to remote work, hybrid setups, and office mandates, most debates miss the real point. It's not about which model is _better_ in some...

  • Apr 25, 2025
    Psychological Safety and Productive Conflict: The Hidden Link Driving High-Performing Teams

    Conflict Isn't the Enemy--Fear Is It's tempting to equate 'healthy teams' with harmony. No arguments, no friction, no tension--just a constant chorus...

  • Apr 25, 2025
    The Rise of Single-Serving Software

    The cost of building has collapsed so far that software can now be disposable on purpose -- and that changes what's worth building.

  • Apr 22, 2025
    The Phantom Projects of Overhiring

    One of the most quietly corrosive things a company can do is overhire. Not because people are malicious or lazy.

  • Apr 21, 2025
    The Confidence Cliff: Why Overcertainty Kills Good Decisions

    You've probably seen this play out. Someone shares an idea--bold, certain, maybe even brilliant-sounding.

  • Apr 21, 2025
    Same Data, Same Dance: Why the Moat Isn't Technical Anymore

    When every security tool starts with the same data, the only real moat is what you do after ingestion.

  • Apr 20, 2025
    The Dehydrated Entity: Hire Only When You're Truly Underwater

    Varun Mohan's hiring philosophy: only add headcount when critical work is actively being dropped. Not deferred. Dropped.

  • Apr 19, 2025
    The Responsiveness Principle: How Small Signals Build Big Trust in Venture

    In early-stage venture, responsiveness is a proxy for dependability. Founders remember how you behaved in the process.

  • Apr 19, 2025
    The Answer Is Obvious -- You Just Don't Like It

    Most 'intractable' problems aren't unsolvable. They just require giving up something you're emotionally attached to.

  • Apr 17, 2025
    AI Detection Hysteria: When Human Creativity Gets Mislabeled

    A photographer friend posted a sunset photo after three hours of waiting for the perfect light. Within minutes: 'Obvious Midjourney.' 'Nice prompt, bro.'

  • Apr 16, 2025
    The Secondary Celebration: A Founder's Perspective on VC Liquidity Events

    During my morning LinkedIn scroll, I came across yet another post from a venture firm celebrating a massive return multiple from a secondary transaction.

  • Apr 14, 2025
    Reading Between the Lines: How to Decode Community Investment Rounds

    When a VC-backed startup pivots to retail funding, you're not getting democratized access. You're getting the deals the wealthy already passed on.

  • Apr 8, 2025
    The Illusion of Traction: When Technical Founders Mistake Interest for Product-Market Fit

    Technical founders consistently confuse polite interest with genuine demand. Here's how to tell the difference.

  • Jan 8, 2025
    The Complexity We Take for Granted

    We live in a world of invisible complexity. Every mundane moment is powered by an intricate dance of systems, protocols, and human ingenuity that we...

  • Jan 7, 2025
    The AI Skill Mirror: Why Technical Interviews Need a Complete Rewrite

    AI doesn't make everyone equally skilled. It amplifies existing ability. That changes what technical interviews should test.

  • Jan 6, 2025
    How RAG Actually Works: Architecture Patterns That Scale

    Deep dive into RAG architectures: chunking strategies, retrieval methods, embedding optimization, and production patterns with research-backed analysis.

  • Jan 6, 2025
    Prompt Engineering Science: I Tested Temperature and Top-P on 1000 Queries

    Systematic experiments on temperature and top-p sampling parameters across 1000 real queries with empirical data on creativity, coherence, and...

  • Nov 27, 2024
    The Product Manager's Secret Superpower: Finding Signal in the Noise

    The best product managers have a superpower that's rarely discussed: they can spot the same underlying user need manifesting in completely different...

  • Nov 25, 2024
    The Three Types of Startup Advice (And Why They're All Wrong)

    The most dangerous thing about startup advice isn't that it's wrong -- it's that it's partially right. Each type comes with its own flavor of wrong.

  • Nov 23, 2024
    The Ubiquiti Effect: Why Enterprise Software Needs a Consumer Revolution

    Ubiquiti proved that enterprise-grade power doesn't require enterprise-grade suffering. Most B2B software hasn't gotten the memo.

  • May 4, 2024
    When the Ask Feels Awkward, It's Already Too Late

    When the Ask Feels Awkward, It's Already Too Late: There's a thing someone on your team is supposed to own. But you hesitate to bring it up.

  • Apr 22, 2024
    Passive-Aggressive Managers Are Organizational Parasites

    Stop making excuses for passive-aggressive managers. They are not misunderstood introverts or conflict-averse leaders. They are parasites destroying...

  • Apr 20, 2024
    Founder-Led Sales: The Art of Selling Your Vision

    For startup founders, sales isn't just another function--it's the lifeblood of your business. Early on, founders are usually the lead salesperson,...

  • Apr 20, 2024
    Performance Reviews: A Guide for Modern Leaders

    Annual performance reviews are a failure theater that helps no one. Here's what actually works.

  • Apr 20, 2024
    Thinking Frameworks: Tools for Better Decision Making

    In the relentless push to build and scale, organizations often overlook a critical piece of infrastructure: how decisions get made.

  • Apr 11, 2024
    When the AI Starts Complimenting You Too Much: A Troubling First for ChatGPT

    OpenAI recently rolled back a GPT-4 update due to sycophantic behavior. The word itself--'sycophantic'--feels like a punchline from a _Black Mirror_...

  • Apr 11, 2024
    AI Expectations: Managing the Hype Cycle

    Most AI products are designed to fail. Not because the technology is bad, but because product teams are building for the wrong expectations entirely.

  • Apr 11, 2024
    Engineering Recognition Through Evals: My Technical Journey Building Shout

    Building an LLM-powered evaluation system to measure whether engineering recognition is specific, accurate, and connected to impact.

  • Apr 11, 2024
    Chrome Extension for Jira Titles: A Developer's Journey

    I kept writing terrible JIRA titles during customer calls. So I built a Chrome extension to fix it.

  • Apr 11, 2024
    Dark Patterns, Bright Lessons: Ethics in Product Design

    The FTC just dropped a 44-page complaint against Uber for deceptive practices around its Uber One subscription. What it reveals about growth culture.

  • Apr 11, 2024
    Engineering and Product Collaboration: Breaking Down Silos

    'This isn't what we asked for.' Five words that strike dread into every engineering team. Five words that signal a fundamental breakdown in the...

  • Apr 11, 2024
    Keep Your Hands Dirty: Why Executives Who Stop Doing The Work Destroy Companies

    The Executive Trap I've seen it happen a dozen times: A brilliant engineer becomes CTO and suddenly decides their job is 'managing the engineering...

  • Apr 11, 2024
    Most PMs Are Too Afraid to Say No

    The product management profession has a cowardice problem. Most PMs will build anything a loud customer demands rather than face an uncomfortable...

  • Apr 11, 2024
    The Homelab That Replaced My Cloud Bill

    I spent $2,000 on hardware that now handles workloads that would cost $500/month on AWS. The cloud is a tax on people who cannot be bothered to learn...

  • Apr 11, 2024
    The Security Tool Comparison Problem

    Every security tool comparison site is funded by the vendors being evaluated. This creates a specific, structural problem for security teams making...

  • Apr 11, 2024
    The Perfection Paralysis: Why Moving Too Carefully Kills Startups

    The most valuable code I've ever written was messy, quick, and written in response to an immediate customer need.

  • Apr 11, 2024
    The Product Manager's Guide to the Perfect Breakfast

    Your company's obsession with meetings is destroying your product team's ability to think. Breakfast is the canary in the coal mine.

  • Apr 11, 2024
    The Abstraction Trap: When Clean Code Goes Wrong

    The most insidious form of technical debt does not come from rushed code or tight deadlines - it comes from overly clever abstractions built too early.

  • Apr 11, 2024
    Quality: The Foundation of Sustainable Growth

    Quality in the wrong places will kill your startup faster than no quality at all. The question is not how much quality, but quality where.

  • Apr 11, 2024
    Don't Build What They Ask For: The Art of Need-Finding

    'If I had asked people what they wanted, they would have said faster horses.' This quote, often attributed to Henry Ford, encapsulates one of the most...

  • Apr 11, 2024
    Your Security Team Cannot Keep Up With AI

    Security review cycles designed for deterministic software are blocking AI adoption. The teams that survive will automate guardrails instead of...

  • Apr 11, 2024
    If You're Serious About Building, You Need to Be in San Francisco

    Remote work is great for lifestyle optimization. It's terrible for building world-changing companies. If you want to build something that matters, San...

  • Apr 11, 2024
    Tech Debt Velocity: Measuring the True Cost of Shortcuts

    The most expensive software I've ever written was code I wrote 'quickly.' Not because it was complex, but because I wrote it with the intention of...

  • Apr 11, 2024
    The Integration Tax: What Nobody Tells You About Building Modern Software

    Every integration you add is a long-term commitment you're not budgeting for. The technical implementation is the easy part.

RSS / Atom

emailgithubx