Archive
Everything,
in one place.
The complete chronological index of ideas, observations, essays, and morning walks published on this site. Search the back catalog by keyword or filter by type.
24 pieces
From Direct Mail to AI: How I Ended Up Building VYNS
I was 12 when I printed 1,000 flyers for a landscaping business and put them in mailboxes around my neighborhood. That instinct never went away. Here is the full story of how I ended up building VYNS.
The Biggest Challenge Isn't Building Smart AI. It's Building AI People Trust Enough to Act On.
Five hard strategic questions about VYNS. Feedback loops, consensus vs. challenger AI, data moats, execution trust, and the founding story. The sharpest insight from this walk: analytical trust and execution trust are different things. Most…
I'm Not Building an AI Tool
The framing I've been using for VYNS is wrong. After an hour of thinking out loud this morning, I realized I'm not building a tool. I'm building a system that helps people decide what to do, and then helps them do it. That's a different cat…
Strategy sessions generate obsolete code with no way of knowing which is which
When you're deliberating, actively researching, and pivoting strategy inside a chat on a large language model, you end up with code or CLI prompts that were generated earlier in the conversation but are no longer relevant. Further down in t…
The Walled Garden Problem
During the process of building my personal website and fleshing out content to publish there and use on social media (for both my personal accounts and VYNS) I realized something. Most of it came from conversations with AI, and none of it c…
Native LLM voice chat doesn't offer transcription, and it should
Native LLM platforms that have built-in voice chat functionality don't offer transcription as a built-in feature or add-on. This could be useful for power users or anyone generating specific outputs from their conversations for media and ma…
Voice Studio
Upload your past writing. The AI learns your sentence structure, vocabulary, pacing, tone, and thematic tendencies, then helps you develop stories, draft chapters in your voice, design covers, format ebooks, and market your work. Not AI wri…
RIAninja
A compliance tracking platform for RIAs: automated annual review scheduling, weighted risk scoring, client compliance dashboards, 90/60/30-day reminders, crypto custody integration, and PDF generation. The commercial wedge into a larger pos…
Startup Ranking Platform
Founders opt in to have their startup ideas and development progress ranked by LLMs by sector and stage. Investors, incubators, and accelerators pay for access to the top-ranked ideas and teams as they emerge, before they raise, before they…
Influencer/KOL Intelligence Market
A marketplace connecting crypto projects with Key Opinion Leaders, built around verified performance data rather than follower counts. Influencers ranked by actual outcomes: did the projects they promoted perform? Did their audience act? Tr…
SphereOS
A reimagined computing interface that replaces app grids, tabs, and folder hierarchies with a zoomable 3D space of interactive orbs. You don't open apps. You navigate a living map of your digital world, guided by an AI consensus engine that…
The Seller-Finance Marketplace
With mortgage rates elevated and conventional financing friction at a decade high, seller financing has re-emerged as a viable path for many transactions. A marketplace to surface, structure, and close seller-financed deals with AI-assisted…
PopCity
Not a dating app. Not an events app. The real-world connection layer for cities, connecting people to things to do, people to do them with, local businesses, services, creators, and economic opportunities. AI-orchestrated, city by city. The…
AetherNet
A proposed protocol layer for post-quantum secure crypto custody, designed for institutions and RIAs managing digital assets as quantum computing threatens current encryption standards. Positioned as a Fireblocks competitor for the post-qua…
Purple Dog Listings
Use AI to help homeowners declutter, stage, and photograph their listings without a traditional agent. The roadmap extended into full FSBO support: offer management, disclosure automation, seller-finance structuring. The idea that occupied…
Cities Powered by Their Own Trash
A systems-level proposal for municipal plastic-to-fuel resilience grids: continuous pyrolysis plants, strategic buried fuel reserves, and neighborhood-scale generation networks that keep power on when the grid fails. A city of 100,000 gener…
Home-Scale Solar Pyrolysis
A detailed engineering breakdown of how the solar thermal concentration principles behind large-scale plastic-to-fuel systems can be miniaturized for household use: parabolic dish concentrator, sealed pyrolysis reactor, condensation train,…
The Prediction Reputation Score
A standardized, public reputation layer for anyone making falsifiable predictions: analysts, influencers, AI models. Think credit scores, but for forecast accuracy. The original idea behind VYNS, and a long-term data infrastructure play in…
Groundwire
A live news network where only people physically present at an event can report what's happening. Traditional news seeds a story. Verified eyewitnesses on the ground provide the real-time updates. Location-verified, credibility-scored, prox…
The hidden cost of context
Anthropic markets the 1M token context window as a capability milestone. And it is one. Being able to feed an entire codebase, a full document collection, or months of conversation history into a single API call is genuinely useful, especia…
The most powerful AI tools are being rationed, and solo builders aren't in the first cohort
Claude Mythos leaked this week. Anthropic confirmed it's real, described it as a 'step change,' and said it's currently being tested with a small group of early access customers. That group isn't me. It probably isn't you either. This is ne…
Projects fragment across chats, and there's no native way to chain them
Everything I am building for VYNS involves multiple parallel workstreams across multiple chats. Product, infrastructure, brand, legal, GTM, marketing, build and deployment prompting, and more. Each lives in a different chat thread, often ac…
AI as technical co-founder is real, but the context window is the bottleneck
I use Claude as my main technical co-founder for the VYNS build along with tools from OpenAI, Gemini, and xAI for various reasons I'll likely get into on my blog one day. The collaboration is genuine and unique. I am able to have the system…
LLM APIs can't read their own chat history, and that's a big gap
I structure my chats in Claude as departments for my startup, VYNS. I have a few chats that persist for personal use, and most other one-off chats I end up deleting after I gather the intel I need. But for VYNS, I keep a few persistent chat…