-
Continue reading →: I Went to Seattle Tech Week and Didn’t Watch a Single Pitch
If you judged Seattle Tech Week solely by the headlines, you’d assume the whole five days were a non-stop blur of founder pitches, funding announcements, and VCs staring at slide decks. I took a different route. I didn’t watch a single pitch. Instead, I spent the week chasing something else—the parts of our industry that usually get buried under startup hype. I wanted to see how we protect our psychological health, use technology responsibly, and honor the human element behind the code. The best sessions of the week weren’t delivered by hype-men shouting from a stage. They were grounded, practical,…
-
Continue reading →: Psychological Burnout in the Age of AI (And Why I’m Writing About It)
The Spark at Seattle Tech Week Recently, I attended a Seattle Tech Week workshop at the Seattle Central Library. It was hosted by Dr. Candice Balluru, Licensed Clinical Psychologist and founder of The Workplace Psychologist PLLC. Her session was titled “Psychological Burnout in the Age of AI.” Dr. Candice laid out a compelling look at how the conversation around burnout is shifting. Her core thesis hit home: technological integration and AI tools are accelerating at a rate that directly collides with fixed human biological limits. It was validating, timely, and deeply needed. Especially in an industry currently racing at breakneck…
-
Continue reading →: The O(1) Harness: Building Production-Grade Automation on a 3B Parameter Budget
My friend Kamila wrote a piece recently about why we overcomplicate agentic harnesses. It made me look at my own server closet. Last week, I watched a 70B parameter model spend twelve seconds trying to format a response because the system prompt told it to dynamically choose between a bulleted list or a numbered list. It wasted thousands of context tokens on self-correction. And it got me thinking about waste. We are currently building AI systems like we have infinite compute and zero latency. We don’t. Especially if we want things to run locally, privately, and reliably. When I set…
-
Continue reading →: Curating Over Coding: The Danger of Extending Too Early
I have a feeling there is a quiet crisis brewing in the software industry. Especially in open source. And it’s all about quality. Generative AI has given us the superpower to broadcast and build ideas at lightning speed. It’s never been easier to turn a personal friction point into a useful, working GitHub repository in an afternoon. You build a lightweight utility that solves your specific problem, or you hop into an established project to fix a bug. It works. Then, your coding agent suggests a few features: “Should we add a Redis caching layer? What about refactoring this entire…
-
Continue reading →: Agent History: CLI Session Reconstructor
Portfolio Project: Agent History (June 2026) Agent History: CLI Session Reconstructor A terminal context recovery system that parses AI agent session history to reconstruct active development states, trace branches, and facilitate instant folder jumps upon SSH connection or login. 🧭 Overview & Problem When juggling multiple client projects, switching environments (e.g. Ubuntu VM for web apps vs. Windows PowerShell 7), or logging back in after a weekend, developers often lose context. Staring at a blank shell prompt asking “Where was I?” represents cognitive drag. Since modern agentic workflows (like pair-programming with the Antigravity CLI) leave detailed local traces, Agent History…
-
Continue reading →: Where Was I? Recovering Terminal Context with Gemini 3.5
I work on a few different projects each week. Because of that, my setup is split. I use two distinct environments: a PowerShell 7 console for Windows tasks, and a dedicated Ubuntu VM for web applications. Every time I open a terminal—especially after a weekend or when switching context between clients—I have to play detective. I sit there staring at the blank prompt, asking: What was I actually working on here? My daily driver coding assistant is the Antigravity CLI running Gemini 3.5. It’s insanely fast—clocking in at about four times faster than Claude Opus right now. Because Gemini is…
-
Continue reading →: “Just a Basic Linux Box”: The Quest for Cheap, Unmanaged Hosting
A person in my community asked a simple, timeless question the other day. “Nothing fancy. Just a basic Linux machine with nothing running on it. Where should I go?” Naturally, a dozen developers immediately jumped in with a wall of opinions. Some suggested setting up full homelabs under their desks. Others started talking about seedboxes. But sometimes, you don’t want to run a server in your closet or manage a complex cluster. You just want a clean terminal, an SSH key, and a public IP to host a small side project, script, or sandbox. And you want it to cost…
-
Continue reading →: Is AI Driving Your Browser? What Will Gemini Refuse To Do?
I wanted to see if I could get my browser to do my chores. Not just search for things, but actually do them. Chrome recently introduced Gemini Auto Browse—a feature designed to let an AI complete multi-step tasks for you on the web. So, I picked two tasks on opposite ends of the spectrum: buying groceries on Safeway.com and automating achievement points on the Google Developer Forums. One worked surprisingly well. The other was a hilarious failure. Task 1: The Safeway.com Grocery Run Groceries are a perfect target for browser automation. It’s a repetitive, tedious search-and-add-to-cart loop that eats up…
-
Continue reading →: DetermiNation: Building a Facebook Fitness Tracker in 2010
Back in May 2010, I posted a quick teaser about a Facebook fitness app I was building. I couldn’t name the client back then. They wanted to remain anonymous until after we launched the app in August 2010. It’s been fifteen years. I think the statute of limitations on that NDA is up. The client was the American Cancer Society. Specifically, their DetermiNation program—an athletic charity training program that helps people run marathons and raise money to fight cancer. I recently found the original spec sheets and some launch emails in my backups. Here’s what it was like building a…
-
Continue reading →: Threads Reader: Local AI Pipeline for Social Automation
“Last night I wrote a program that downloads all my posts then used a local inference model to tag the sentiment and then put it in Google sheets to make a bar chart.” That was my Threads post on March 14, 2026. It was a quick summary of a weekend project, but it got me thinking about how easy it’s become to build local AI pipelines. I wanted to see if I could automate reading and replying to Threads posts without sending my data to a cloud API. It turns out you can do it pretty easily using local models…
Hello,
I’m Aaron
Engineering Manager in Seattle
