Kenji

[Kenji @Kenjiding]

阿德莱德招聘-国际跨境物流公司招聘末端物流分析员

欢迎Marketing ，理工科的毕业生申请，有物流经验，跨境电商专业更佳

[Kenji @Kenjiding]

@astro_reid @NASA @NASAArtemis Is this Australia area in this picture ?

[Kenji @Kenjiding]

@ai_shangshu @fkysly 我觉得现在的vide coding 只是一个强力的辅助，会让你效率会提高很多，但是当你试图停止思考，停止理解整个项目，那么ai写的代码也会成为你以后的债务。so never stop thinking

[Kenji @Kenjiding]

@ai_shangshu @fkysly 其实就算你的项目是从头就由vibe coding 去写，但是你依然没有审查，调试，理解的话。后面当出现了一些很难用自然语言描述清楚的bugs，而你又不熟悉代码没办法提供恰当的代码位置和片段，这个时候Ai就可能会“偷懒”随便打补丁修复，久而久之，补丁叠补丁就成了屎山

[Kenji @Kenjiding]

Initial Virtual Computer in sandBox implementation: The pipeline is API/Playwright → Socat → Chromium CDP. Currently running into issues with real-time forwarding via Socat.

#ai #aiagent #llm

[Kenji @Kenjiding]

I'm building a general-purpose AI agent similar to Manus. Currently, I've set up a basic sandbox to execute risky commands securely. The next step is to implement browser automation via the Chrome DevTools Protocol (CDP).

[Kenji @Kenjiding]

@hex_agent Every Agent system has its own unique logic and strategy. You can't just blindly assume 0.8 is your system's optimal threshold; it must be derived from your own extensive data and evaluation

[Kenji @Kenjiding]

Vertical AI Agent for E-commerce Customer Support

This demo presents a vertical AI agent designed for e-commerce customer support workflows.
By integrating a local RAG pipeline with domain-specific knowledge bases, the agent can accurately retrieve..
#ai #aiagent

[Kenji @Kenjiding]

@hex_agent Many people mistakenly treat thresholds as a universal standard, like a magic number of 0.8. But there’s no one-size-fits-all benchmark.

[Kenji @Kenjiding]

@hex_agent logs are so critical. If you're chasing maximum precision, you should have your team manually tag about 300 logs to build a gold standard dataset. Once you visualize the accuracy curve in your eval system, you'll be able to identify the threshold that works best for your setup

[Kenji @Kenjiding]

@hex_agent You can't nail the threshold right away. I’m sticking to a 'Tight-to-Loose' approach: prioritize human hand-offs initially to keep trust high, then use post-mortem log analysis to optimize the triggers based on real-world data

[Kenji @Kenjiding]

@hex_agent Subsequent turns then reference this structured state rather than the 3000-word raw document.

[Kenji @Kenjiding]

@hex_agent don't cache entire messy context, which is noisy and token-heavy, for example I enforce a structured schema (Schema A) for key entities like Order IDs and Return Policies. The AI parses the 3,000-word RAG results and populates this schema A.

[Kenji @Kenjiding]

@hex_agent We would rather terminate the conversation and transfer it to a human agent than allow the AI to generate illusions.

[Kenji @Kenjiding]

Using Rust to power complex frontend features exceeded my expectations.

[Kenji @Kenjiding]

@Menghuanlangqi2 这个说实话我就不太相信了。凤姐

[Kenji @Kenjiding]

@Menghuanlangqi2 致敬英雄！

[Kenji @Kenjiding]

@Menghuanlangqi2 国内的肺结节人数增加很多了，疫苗之父被抓，哎，我也打了