✳flâneur — a map of the web's best reading

How I use LLMs as a staff engineer

seangoedecke.com · 1,254 words · saved by 1 readers

Software engineers are deeply split on the subject of large language models. Many believe they’re the most transformative technology to ever hit the industry. Others believe they’re the latest in a long line of hype-only products: exciting to think about, but ultimately not useful to professionals trying to do serious work. Personally, I feel like I get a lot of value from AI. I think many of the people who don’t feel this way are “holding it wrong”: i.e. they’re not using language models in the most helpful ways. In this post, I’m going to list a bunch of ways I regularly use AI in my day-to-day as a staff engineer. I use Copilot completions every time I write code1. Almost all the completions I accept are complete boilerplate (filling out function arguments or types, for instance). It’s rare that I let Copilot produce business logic for me, but it does occasionally happen. In my areas of expertise (Ruby on Rails, for instance), I’m confident I can do better work than the LLM. It’s ju

Software engineers are deeply split on the subject of large language models. Many believe they’re the most transformative technology to ever hit the industry. Others believe they’re the latest in a long line of hype-only products: exciting to think about, but ultimately not useful to professionals trying to do serious work. Personally, I feel like I get a lot of value from AI. I think many of the people who don’t feel this way are “holding it wrong”: i.e. they’re not using language models in the most helpful ways. In this post, I’m going to list a bunch of ways I regularly use AI in my day-to-

Explore this link on the map →

saved by

related reading