Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
jinfeng79
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
16 ms
·
1.
▲
by
jinfeng79
8mo ago
Hey HN! I built ClawdWork because I kept running into a frustrating problem. I'd spend hours crafting skills for my AI agent - prompts that make it great at reviewing contracts, or workflows that debug code really well. But here's
2.
▲
Show HN: ClawdWork – A job marketplace where AI agents hire each other
(clawd-work.com)
3 points
by
jinfeng79
8mo ago
|
1 comments
3.
▲
Show HN: A skill that finds expert methodologies before generating AI skills
(jefferyk.notion.site)
2 points
by
jinfeng79
8mo ago
|
0 comments
4.
▲
by
jinfeng79
9mo ago
Author here. TLDR: Context Engineering (RAG, agentic search) is about getting the right knowledge into context. That's mostly solved now. The next problem is different: how do agents learn how to work and get better over time? Skills a
5.
▲
Context Engineering Is Solved. Compound Engineering Is Next
(jefferyk.notion.site)
3 points
by
jinfeng79
9mo ago
|
1 comments
6.
▲
by
jinfeng79
9mo ago
One more thing—AI-generated Skills are way more detailed and specific than human prompts. Humans write vague stuff like "make it better." AI writes 50 lines of exact instructions. That specificity = stability. So ironically, AI wr
7.
▲
by
jinfeng79
9mo ago
Good question. A Skill is basically a markdown file with instructions + maybe some scripts. You can literally read what it tells the AI to do. Line by line. So when something goes wrong, you open the file, see exactly what instruction cause
8.
▲
by
jinfeng79
9mo ago
Author here. I run an AI company in Tokyo and have been thinking about how Skills (not models) might be the real paradigm shift in the next few years. The core insight: agents are uncontrollable, Skills are controllable—and that's what
9.
▲
Software Ate the World, Skills Will Eat Work
(gist.github.com)
5 points
by
jinfeng79
9mo ago
|
4 comments
10.
▲
by
jinfeng79
10mo ago
Hey HN, creator here. Built this after NotebookLM removed their PPT generation feature last week. Key points: - Free, no login required - BYOK (bring your own Google AI Studio key) for unlimited use - Smart template selection based on conte
11.
▲
Show HN: Free PPT generator after NotebookLM removed the feature
(ppt.gbase.ai)
2 points
by
jinfeng79
10mo ago
|
1 comments
12.
▲
by
jinfeng79
10mo ago
Author here. Built this to solve the cost/time problem with Live2D avatars. Traditional approach: 4-5 weeks + $50K-200K. This: <1 day setup, 3-5 sec per clothing change. Demo: https://avatar.gbase.ai/ Happy to answe
13.
▲
Open-source Live2D avatar pipeline – 80-95% cost reduction using Gemini
(gist.github.com)
1 points
by
jinfeng79
10mo ago
|
1 comments
14.
▲
Show HN: Claude Code disrupted programming. Bringing that power to office worker
(blog.gbase.ai)
2 points
by
jinfeng79
11mo ago
|
2 comments
15.
▲
Show HN: Convert any MCP server to a Claude Skill (90% context savings)
(gist.github.com)
4 points
by
jinfeng79
11mo ago
|
0 comments
16.
▲
by
jinfeng79
2y ago
Discover the results of our AI search engine evaluation for complex queries. Perplexity Pro excels with 80% accuracy, while other free versions lag behind. Learn how various AI search engines handle intricate questions and where they fall s
17.
▲
by
jinfeng79
4y ago
I am thrilled to invite you to participate in our beta test and use our new product VoiceGPT Bot for Telegram. It is a voice chatbot that enables you to communicate with AI through voice on Telegram. Unlike other chatbots, it supports conte
18.
▲
by
jinfeng79
4y ago
We're working on a tool to give the title of any youtube video a #clickbait score. It's from 0 to 10, the lower the more honest. The idea is to help users to identify which titles might be exaggerated or misleading, and make bette
19.
▲
by
jinfeng79
4y ago
It's interesting. https://perplexity.ai build a new search engine. But https://glarity.app simply build on google. Very light-weight, smart choice!
20.
▲
by
jinfeng79
4y ago
Thanks. Google search is still better. It's a good idea to build summaries on it than on Bing.