HackerTrans
トップ新着トレンドコメント過去質問紹介求人

justanotheratom

513 カルマ登録 15 年前
Working on https://www.kincalendar.com/, https://ingredicheck.app/

Github: https://github.com/justanotheratom

Past: www.linkedin.com/in/realsanketp

投稿

Which Predictor? How GEPA Optimizes a Multi-Agent DSPy Program

elicited.blog
2 ポイント·投稿者 justanotheratom·一昨日·1 コメント

You won't be talking to AI in natural language for too long

elicited.blog
5 ポイント·投稿者 justanotheratom·先月·8 コメント

Teaching Codex to Test a Voice-First Calendar App

elicited.blog
1 ポイント·投稿者 justanotheratom·2 か月前·1 コメント

[untitled]

1 ポイント·投稿者 justanotheratom·3 か月前·0 コメント

Is Data Curation the New Feature Engineering?

elicited.blog
1 ポイント·投稿者 justanotheratom·7 か月前·1 コメント

Waiting for the Let's Encrypt Moment for Content Authenticity

elicited.blog
2 ポイント·投稿者 justanotheratom·7 か月前·0 コメント

Strategy and Program: Two Concepts for Working with DSPy

elicited.blog
1 ポイント·投稿者 justanotheratom·7 か月前·0 コメント

Non-Obvious Things I Learned About GEPA

elicited.blog
2 ポイント·投稿者 justanotheratom·7 か月前·1 コメント

DSPy Parallel Chunk Streaming

elicited.blog
3 ポイント·投稿者 justanotheratom·7 か月前·0 コメント

コメント

justanotheratom
·一昨日·議論
Appreciate anyone pointing out technical issues with this article.
justanotheratom
·15 日前·議論
Fair.

There are, however, people who have to seriously avoid certain ingredients, and it is a pain to go through all the fineprint, which is sometimes made quite difficult to read. For example, check the Asian section in your grocery store.

If you are shopping for multiple people in the family, some of which have some restrictions, it is hard to map everything confidently when shopping.

If you are traveling abroad, like Japan for instance, especially with dietary restrictions, an app like this can be quite handy.
justanotheratom
·先月·議論
with skills, model can easily learn the jargon.
justanotheratom
·先月·議論
exactly. to mind-meld with your AI, you will have to use/understand quite dense language.
justanotheratom
·先月·議論
Would love some critique of my thoughts.
justanotheratom
·2 か月前·議論
How I used Codex, FlowDeck, and Loopback to automate testing of Voice Interactions for my Voice-First App: https://www.elicited.blog/posts/teaching-codex-to-test-a-voi...
justanotheratom
·7 か月前·議論
I am not an expert in either ML or AI, so appreciate expert feedback on my thoughts and learnings.
justanotheratom
·7 か月前·議論
haven't succeeded at GEPA yet..
justanotheratom
·昨年·議論
F# was my favorite language, but - You have to chose the language of the Domain that you are working in, e.g, Swift for native iOS Development. Supabase Backend requires TypeScript, etc. - LLMs don't care about F#.