Dev log #18 From Redis Brokers to Knowledge Graphs: A 177-Commit Sprint
Hit a perfect 7-day streak this week, shipping 177 commits across systems architecture and personal...
找到 3 篇相关文章
Hit a perfect 7-day streak this week, shipping 177 commits across systems architecture and personal...
I still remember the spreadsheet. Keyword, search volume, difficulty score. Pick the keyword, hit the target density, write the meta description, done. I built entire client strategies around that sheet for years, and it worked well enough that I never questioned it too hard. I opened an old version of that spreadsheet last month while helping a client plan content for the fall. Half the columns didn't matter anymore. Not in a dramatic, the-sky-is-falling way. Just quietly, the way a tool stops getting used and you don't notice until you go looking for it. The keyword still matters, the strategy around it doesn't I'm not going to tell you keywords are dead. People still type words into search boxes, and Google still uses them to figure out what a page is about. That part hasn't changed. What's changed is everything I used to do around the keyword. Density checks feel almost silly now. Nobody's counting how many times "best content management system for small business" appears on a page, least of all an AI model summarizing five sources into one paragraph. It's reading for meaning, not repetition. Stuff a keyword in five times and you're not helping your odds, you're just writing worse. I used to treat the first 100 words as prime real estate for the primary keyword. These days I treat them as prime real estate for the actual answer. Those aren't always the same sentence anymore, and that distinction is doing a lot of work I didn't used to think about. Backlinks matter less For most of the last decade, if you'd asked me the single highest-leverage thing a freelancer could do for a client's SEO, I'd have said links without much hesitation. Get mentioned somewhere credible, get linked from a real site in your niche, and rankings would follow eventually. Links still count for something. I'm not throwing that out. But I've watched pages with a thin backlink profile show up in AI-generated answers ahead of pages that would've dominated the old rankings, purely because the
Of all the things involved in turning a woman's life into a 60-second reel, I assumed that the writing would be the easy part. Surely telling a good story in 150 words is exactly what a large language model should be good at; yet it proved surprisingly difficult. Draft after draft suffered from common AI weaknesses: a tendency to use hyperbole and inspirational language, to generalize, follow generic founder arcs ("built in a basement"), and focus on morbid details. (For this effort, I was using Sonnet 4.6, which I found to be better — more grounded, more true to facts, less inventive — than Opus 4.7 at creating longer bios.) Getting to something publishable took a significant amount of work. That said, the human in this story also found writing good short story arcs surprisingly challenging, so the effort spent on getting the system to do it decently was well worth it. Here's some context, then what I did. I've been publishing short biographies of notable women for a while now, on a website called Tycoona . Why notable women? Because in each field there are so many women who contributed so much and are little known. I've never understood why Corita Kent , the pop-artist nun, isn't as famous as Andy Warhol, why Hetty Green , the Gilded Age value investor called both the "Witch of Wall Street" and the "Queen of Wall Street," disappears into history behind Benjamin Graham and his protege Warren Buffett. The problem I faced is that no one was seeing my bios, so I decided to make short videos or reels in hopes of increasing my reach. Creating the technical infrastructure for the reels on top of my existing system was fun and relatively straightforward. Each reel consists of a series of "beats," and Remotion turns those beats into a vertical reel, complete with on-screen text and royalty-free images & attributions pulled from Wikimedia, Flickr, or Library of Congress. For content, the beat generator relies on the knowledge base of validated facts that my system creates f