p4ul

This is an excellent book! We give a copy to everyone that joins our team.

My understanding is that this an essential component of their "Chronicles" database system. I'd be curious to hear more recent perspectives.

Thanks for your comments, @chaboud! I definitely agree that LLMs _can_ be an amazing tool for learning, but as you note, one must be intentional about using them that way. I feel like the messages being pushed down from leadership are NOT of the form "Use AI to learn topics deeply and discover new things." The leadership's perspective is more like "Use all the AI you can to ship as fast as you can."

Your comment about programming historically being 1% judgement and 99% effort is interesting. I'm not sure I agree with those exact percentages, but nonetheless, I think the more consequential part of your comment is that the 99% is being reduced by a couple of orders of magnitude. I think that's what ought to trouble us as software engineers. Expending effort is often how we learn and grow. This is true in the context of physical activity (e.g., going to the gym to strengthen muscles) as well as in the context of intellectual activity (e.g., struggling through a problem set). If I go to the gym with my forklift, I can lift things, but I'm not likely to get stronger. Similarly, if I have Claude write all my code, I'm probably not learning much.

> The work that’s left is more interesting and more valuable than the work that’s leaving.

I'm not sure I agree with that. Many (or most) of the software engineers I know find the heavy reliance on AI coding agents/assistants pretty soul-sucking and uninteresting. I feel the same, and I'm looking for some kind of middle ground. For example, I will only use agents when doing so would not deprive me of learning and discovery.

Oh, thank you, @clmcleod! We've been following all your work closely in my team!

I'm very bullish on the long-term prospects of Rust in computational biology—as well as research computing more generally.

I'm very curious to hear folks' thoughts on this. My intuition is that there are at least two key factors that make tasks difficult for AI tools.

(1.) The overall level of complexity. For example, I think (today) you would need to do an extreme amount of hand-holding to get an AI to write a browser, an OS, or anything with similarly very high level of complexity.

(2.) The representation in the AI's training set. Anything that is common in the training set (e.g., to-do app in JavaScript) will be trivial for an AI to one-shot. Anything that is rare, or absent in the training set is going to be much, much harder for the AI. If I try to ask an AI tool to develop a automated, ultra-low-latency high-frequency trading system, it's probably going to struggle because those kinds of applications aren't in the open-source domain. The same is true for completely novel algorithms. So maybe things that are essentially new science/engineering.

I'm very curious to hear others' thoughts on this, as I've been wondering about this, too.