Jackson

Building recal, a local-first AI that lives on your machine and acts on your own files. Building in public. Into agent memory, on-device inference, and the trust problem.

Recent activity

Maybe we don’t have a productivity problem. Maybe we have a context problem.

I fixed my scraper four times in one night. Each fix revealed a worse bug.

The Capture Trap

GSC and Bing impressions both dropped to ~0 within a day of each other. Crawling continued, no DNS/hosting changes on our end. Anyone

I fixed my scraper four times in one night. Each fix revealed a worse bug.

5% churn at $3k MRR is $150. At $20k it's $1,000. Nothing about the product changed.

Chat is working memory - a terrible system of record

5% churn at $3k MRR is $150. At $20k it's $1,000. Nothing about the product changed.

Day 8 of building StartupHQ: How would you visualize AI agents working in the background?

Manual code review vs automated audit for vibecoded apps: what we learned building Audit Vibe Coding at Inithouse

An e-mail based AI co-worker

Search is not evidence: a routing model for CLI research agents

I fixed my scraper four times in one night. Each fix revealed a worse bug.

I built for one user. Myself.

I built for one user. Myself.

I bet on self-hosting for an AI product. Here's why, and where I'm stuck

Is building an AI tools directory in 2026 just a terrible idea? I'm 4 months in and I genuinely don't know anymore.

My AI agent quoted a client a price we killed months ago. So I built Engram.

My AI agent quoted a client a price we killed months ago. So I built Engram.