TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Winner of the SF Mistral AI Hackathon: Automated Test Driven Prompting

95 pointsby carlcortrightabout 1 year ago

8 comments

bugglebeetleabout 1 year ago
Isn’t this just a very naive implementation of what DsPY does?<p><a href="https:&#x2F;&#x2F;github.com&#x2F;stanfordnlp&#x2F;dspy">https:&#x2F;&#x2F;github.com&#x2F;stanfordnlp&#x2F;dspy</a><p>I don’t understand what is exceptional here.
评论 #39842566 未加载
hitchstoryabout 1 year ago
I built something like this too:<p><a href="https:&#x2F;&#x2F;github.com&#x2F;hitchdev&#x2F;hitchstory&#x2F;blob&#x2F;master&#x2F;examples%2Fllm%2FREADME.md">https:&#x2F;&#x2F;github.com&#x2F;hitchdev&#x2F;hitchstory&#x2F;blob&#x2F;master&#x2F;examples%...</a>
k__about 1 year ago
Just yesterday, I wrote an article about FT and learned about services like Entry Point AI.<p>Seems like an awesome idea. I&#x27;m curious how long it will take to get a model on a reasonable level.<p>Phind is pretty good and also the fastest model I used recently, so I&#x27;d assume it&#x27;s quite small, no?
carlcortrightabout 1 year ago
Post Author: getting a lot of requests so scaling the backend. Standby.
评论 #39843292 未加载
toisanjiabout 1 year ago
can you please add more info on the page to show why it is important and how its helpful
评论 #39843348 未加载
imranqabout 1 year ago
Shouldn&#x27;t this be LoRA training?
paraditeabout 1 year ago
Yeah I&#x27;ve been thinking about this lately.<p>LLMs come and go.<p>Prompt engineering techniques come and go.<p>But eval &#x2F; labelled dataset is always useful once you built it.
评论 #39842446 未加载
评论 #39842878 未加载
carlcortrightabout 1 year ago
“Learning” by prompting, calculating the loss against evals, and updating the prompt
评论 #39842881 未加载