Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
samchon
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
1.
▲
Function Calling Harness: Success Rate From 6.75% to 100%, by compiler skills
(typia.io)
3 points
by
samchon
6mo ago
|
1 comments
2.
▲
by
samchon
6mo ago
I was personally invited by the Qwen team to speak at Qwen Meetup Korea, and got to present locally here in Korea yesterday — pretty honored to have been reached out to directly. The talk was about how I got function calling to work reliabl
3.
▲
by
samchon
6mo ago
Next subject: function calling for domains where the final answer is unverifiable. Legal rulings, business decisions: you can't verify the conclusion. But you can verify the process. Model every mandatory reasoning step as typed schema
4.
▲
Show HN: Qwen Meetup Presentation, Function Calling Harness, 6.75% to 100%
(autobe.dev)
2 points
by
samchon
6mo ago
|
1 comments
5.
▲
by
samchon
6mo ago
Just allowing Yolo, and sometimes do rolling back
6.
▲
Show HN: Qwen Meetup, Function Calling Harness, turning 6.75% to 100%
(autobe.dev)
1 points
by
samchon
6mo ago
|
0 comments
7.
▲
by
samchon
10mo ago
Amazing
8.
▲
by
samchon
1y ago
I came in thinking it was a free ebook lol
9.
▲
by
samchon
2y ago
Challenging hard work. Hope to go well.
10.
▲
by
samchon
2y ago
I'd used until 2020, and moved away to Windows laptop. The key reason of moving is that, Apple suddenly moved Korean/English change to another position. In Korea, the language changing key is positioned at the right of space. And
11.
▲
by
samchon
2y ago
Great insight making me think a lot
12.
▲
by
samchon
2y ago
I am not a financial trader, but I had a similar experience while building a chatbot that analyzes the economy and current affairs. I focused on function calls and secured as many papers, news, and statistical APIs as possible (and wrapped
13.
▲
by
samchon
2y ago
I think CRUD API may remains (called by chatbot), but CRUD app may not. I got this insight making a shopping mall chatbot from a backend server. It has 289 API functions, and all of them are properly called in the AI agent. With this experi
14.
▲
by
samchon
2y ago
My github account name is samchon, and cannot find my principle repositories like typia from your website
15.
▲
by
samchon
2y ago
I wrote some of my libraries, but no reaction at all from the website. It's sad that I'm not famous ToT
16.
▲
by
samchon
2y ago
If there are some prepared functions, I do not provide it as tools directly. Instead, make the orchestration to pass a function seleting agent. This strategy makes success rate of function calling higher. Github: /wrtnlabs/agentic
17.
▲
by
samchon
2y ago
The background is yellow and it hurts my eyes ToT
18.
▲
by
samchon
2y ago
This is interesting, take my star
19.
▲
by
samchon
2y ago
Can't I do function calling in OpenAPI? I also feel like MCP is reinventing the wheel. I have been converting OpenAPI documents into function calling schemas and doing tool calling since function calling first came out in 2023, but it&
20.
▲
by
samchon
2y ago
Our framework concentrates on DX (developer experience) by TypeScript compiler skill. I guarantee you that this is the easiest AI framework in the world. As our framework and website domain are not famous, so if I direct link to the homepag
21.
▲
Show HN: Agentic AI framework specialized in DX and function calling
(github.com)
8 points
by
samchon
2y ago
|
2 comments