For example, you can spend a few hours writing a really good set of initial tests that cover 10% of your codebase, and another few hours with an AGENTS.md that gives the LLM enough context about the rest of the codebase. But after that, there's a free* lunch because the agent can write all the other tests for you using that initial set and the context.
This also works with "here's how I created the Slack API integration, please create the Teams integration now" because it has enough to learn from, so that's free* too. This kind of pattern recognition means that prompting is O(1) but the model can do O(n) from that (I know, terrible analogy).
*Also literally becomes free as the cost of tokens approaches zero
A neat part of this is it mimics how people get onboarded onto codebases. People usually aren't figuring out how to write tests from scratch; they look at the current best practices for similar functionality in the codebase and start there. And then as they continue to work there they try to influence new best practices.
FYI, you're missing two patterns that allow the `--key=value` admirers and the `-alltheshortopsinasinglestring` spacebar savers among us to be happy (for the otherwise excellent options parsing code).
shopt -s extglob
case "$1"
# Flag support - allow -xyz z-takes-params
-@(a|b|c)*) _flag=${1:1:1}; _rest=${1:2}; shift; set -- "-$_flag" "-$_rest" "$@";;
# Param=Value support
-?(-)*=*) _key=${1%%=*}; _value=${1#*=}; shift; set -- "${_key}" "$_value" "$@";;
esac
It sends a flag that dangerously allows Claude to just do whatever it wants and only give us the final answer. It doesn't do the back-and-forth or ask questions.
Looks great! My website has a /life page (https://anandchowdhary.com/life) where I track all my life & health data, including:
- yearly themes and quarterly personal OKRs
- my live location (yes, really)
- books I read, music I listen to
- biomarkers, health and fitness data, sleep records
Cool page! Maybe you can ask some data brokers what interoperability standards they use so you can provide the correct file to them, perhaps even negotiate a good price for the data ;)
Haha indeed... if it helps, I built https://stethoscope.js.org where I used official & unofficial APIs and takeout exports to compile everything in one place.
Haha that's true. I do only store the geolocation up to 2 decimal places so it's rounded a little, but people do find out when I leave town. Luckily I have camera/alarm systems/etc. but maybe my insurance will tell me I brought this on myself. I even had https://x.com/anandstalker live-tweeting it before Twitter made their API too expensive.
The overwhelming majority of burglars are not doing online reconnaissance to establish where one person might be, when you can just drive by and see if there are cars parked there, or just kick in the door and see if anyone yells.
Yes and no. It’s indeed a playground to test out various models and gives you an endpoint to play with it, but it’s not that developers can upload their own custom models. Instead, it’s currently only a curated library of certain popular models like those from OpenAI, Microsoft, and Meta.
(I don’t work at GitHub but was quoted in the article).
Growing up in India, we learned that the four pillars of democracy are the Legislature, the Executive, the Judiciary and the Media. The nonprofit presumably is saying that they/nonprofits are the fifth pillar.
Here are receipts some from 2020: - https://github.com/AnandChowdhary/bsc-thesis - https://github.com/AnandChowdhary/slack-netlify-trigger - https://github.com/AnandChowdhary/analytics-icons