Re:TSMC execs allegedly dismissed OpenAI CEO Sam Altman as '...
by jeremy151 on Friday September 27, 2024 @10:08AM
(#41670691)
I find tools where I am manually shepherding the context into an LLM to work much better than Copilot at current. If I think thru the problem enough to articulate it and give the model a clear explanation, and choose the surrounding pieces of context (the same stuff I would open up and look at as a dev) I can be pretty sure the code generated (even larger outputs) will work and do what I wanted, and be stylistically good. I am still adding a lot in this scenario, but it's heavier on the analysis and requirements side, and less on the code creation side.
If what I give it is too open ended, doesn't have enough info, etc, I'll still get a low quality output. Though I find I can steer it by asking it to ask clarifying questions. Asking it to build unit tests can help a lot too in bolstering, a few iterations getting the unit tests created and passing can really push the quality up.
I'm probably one of those "heavy users", though I've only been using it for a month to see how well it does. Here's my review:
Large completions (10-15 lines): It will generally spit out near-working code for any codemonkey-level framework-user frontend code, but for anything more it'll be at best amusing and a waste of time.
Small completions (complete current line): Usually nails it and saves me a few keystrokes.
The downside is that it competes for my attention/screen space against good old auto-completion, which costs me productivity every time it fucks up. Having to go back and fix identifiers in which it messed up the capitalization/had typos, where basic auto-complete wouldn't have failed is also annoying.
I'd pay about about $40 right now because at least it has some entertainment value, being technologically interesting.