Measurement Why Measurement Beats Optimization (And Why I Keep Having to Remind Myself) About six months into building COS, I spent a week rewriting the output format for one of the scoring dimensions. The reports looked cleaner. The language was tighter. I felt good about the work. Then I realized I had no data on whether users were actually reading that section of
Build In Public The Next Build Is Already Inside This One KnowledgeForge could route a reasoning task to the right mode. What it couldn't do was tell me if the output worked for a specific reader. That gap showed up the first time I tried to use it to evaluate outreach copy. I had a cold email. KF could
Build In Public The Maintenance Bet Open-sourcing something isn't a technical decision. It's a promise about your future attention. Most people treat it like a launch. You clean up the repo, write a README, push public, post the link. That's the visible part. The invisible part is the commitment
Build In Public The Assumption You Didn't Know You Had The cold reader doesn't reveal that you wrote badly. They reveal the assumptions you forgot you'd made. I asked people to read the KnowledgeForge README and tell me where they got lost. Someone came back with a specific question: "What does it mean for a
Build in Public Private Tools Don't Export Three years of private notes are harder to explain than I expected. What writing the README taught me about whether the compression was real.
KnowledgeForge Open Source — One Week In KnowledgeForge went open source last week. Three years of building it in private, 43 modules, a forced rationalization, and what it became.
Marketing How I Think About Audience Personas Differently Now For years I thought about audience the same way everyone teaches it: build the persona. Job title, company size, industry, pain points, maybe a stock photo of someone named "Michelle the Marketing Director." Put it on the wall. Write to Michelle. It's not wrong. It'
Marketing The Framing Problem in Technical Marketing The first sign I had that something was off was a conversation that went nowhere. I'd spent about twenty minutes explaining how COS worked to someone who ran demand gen at a mid-size B2B software company. Measurement against peer-reviewed frameworks. The OCEAN model. Five dimensions of
KnowledgeForge 29 Versions, 23 Failures, and the Three-Line Fix That Changed Everything Six weeks ago I asked a simple question: could an AI system redesign a battle bot, test it in a live competitive arena, diagnose its own failures, and improve itself — without any human involvement in the loop? The answer turned out to be yes. It also took 23 consecutive failures to get there.
Automation Why I Track What I Got Wrong More Carefully Than What I Got Right Nightwatch has been running overnight tasks for about six months. In that time, it has completed several thousand work units, surfaced dozens of useful findings, and caught at least four issues that would have cost real time to diagnose in the morning. I track all of this in a log.
AI Tools What I Learned Building Solo AI Tools: On Limits and Leverage Solo founders building AI tools hit limits that are structural, not effort-related. Working harder doesn't fix them. Three limits and three leverage points — the ones that actually matter.
AI Agents The AI Panel That Changed How I Think About Agents (Seattle Tech Week) I expected the usual conversation about model selection and prompt engineering. What I got was a room full of builders talking almost exclusively about failure modes and environment design.
B2B Communications Which Cialdini Principle to Lead With for Each B2B Buyer Type Originally published on semalytics.com/blog. Cialdini's six principles work. The problem is knowing which one to lead with for the specific buyer in front of you. In B2B, getting that wrong isn't neutral. It costs the deal. Most B2B marketers know the principles. What they
AI Tools Nightwatch: What I Built When I Needed Someone to Work While I Sleep I had a list of monitoring and research tasks that never got done. Not because they weren't important — because they required sitting at a computer. So I gave a machine the keys.
B2B Communications The B2B Cold Email Personality-Fit Framework: Write to the Role, Not the Average Prospect Originally published on semalytics.com/blog. Most cold email advice is about getting opened. That's not where the deal dies. The deal dies in the body copy — after the open — when the message is calibrated for a generic prospect instead of the person reading it. The B2B cold
Build in Public The Scores Improved. The Results Got Worse. I built a tool to grade my writing. Then I wired it into the thing it was grading.
B2B Communications How to A/B Test Email Subject Lines by Personality Type Originally published on semalytics.com/blog. Most email A/B tests are designed to find a winner. The problem is that the winner is an average across people who wanted completely different things — and the test can't tell you which variant worked on whom. The fix is not
B2B Communications Personality-Based Marketing for B2B SaaS: The Signal Your Segments Are Missing Originally published on semalytics.com/blog. Personality-based marketing for B2B SaaS means calibrating your copy's frame, language, and appeals to the buyer's Big Five trait profile — not just their job title or plan tier. A VP of Engineering with high Conscientiousness evaluates risk differently than
AI Coding FORGE: What Happens When You Spec Before You Build PAIRED solved multi-agent execution. The bottleneck moved to the spec. FORGE automates decomposition: natural language in, explicit task graph out.
B2B Communications Audience-Fit Scoring for B2B Communications: The Pre-Flight Check A/B Testing Can't Do Originally published on semalytics.com/blog. Audience-fit scoring measures how well a piece of B2B copy matches the specific profile of its intended audience across four dimensions: personality traits, emotional engagement, message clarity, and cognitive framing. The score runs before the copy ships. Not after. Before. COS (Communications Optimization
B2B Communications Why B2B Marketers Are Moving from Demographics to OCEAN Personality Scoring Originally published on semalytics.com/blog. The OCEAN model (also called the Big Five) identifies five personality traits that predict how B2B buyers process and respond to copy. Applied to B2B marketing, it answers a question that demographic targeting can't: not just who is reading, but how they
AI Copywriting How to Write Prompts for AI Copy That Sounds Like You Originally published on semalytics.com/blog. Your team drafts a LinkedIn post in ChatGPT. Everyone reads it, everyone nods. Nobody argues it's bad. Nobody says it sounds like you either. You edit it. You push some sentences around. You add a specific number. You cut the third bullet.
Build in Public Why I Stopped Using One AI Agent for Everything I fell in love with Windsurf for about three months. The agentic coding experience was different. More fluid than Cursor. It felt like the assistant was in the project with you, not just answering questions about it. Then I started building things complex enough to expose the problem. Long sessions,
Build in Public Three Years at AI Tinkerers. Three Presentations. One Pattern. I've been showing up to AI Tinkerers for three years. I started going before I had anything worth demoing. I kept going after I did. The difference between those two states taught me more than the presentations. AI Tinkerers is a monthly meetup for people building with AI.
