Agents are increasingly being trusted to do quantities of work that cannot be verified at code-level detail. Finding verification criteria for the products of work done like this will be critical for...
I've been playing around with using Voice Activity Detection to record multiple separate audio channels, then stitching those together into a transcript with Whisper to do speech to text.
I'm working through my second implementation of an idea that has been bouncing around my head for a while. The general idea is to be able to transcribe a conversation in real time and have an agent...
I'm not sure if they originated this pattern but I like how Claude Code introduces new features with tips.
Notes on way too many hours spent agent coding over the holidays.
I am working on adding a feature to Tarn where I can type @@ and it will pop a list of content in my blog I can link to in markdown style. This is an idea borrowed from Obsidian for easily reference...
After hitting my head against a few different open source Swift text editor implementations, I ended up settling on a webview with a Monaco in the Swift app. The syntax highlighting isn't exactly how...
I've been working with Claude Code to build a tool to help me write in this blog. I've historically used Cursor to write in MDX and Claude Code to help build and wire up custom Astro and sometimes...
I've seen coding agents refuse or avoid writing comprehensive tests because they claim "most tests are passing" or "90% is great coverage" even when prompted to do something like "fix the tests" or...
I probably would have switched off of Nix for system management, but with Claude Code, any problems I run into are as good as solved.