@zhanghaili0610: Just wrapped my talk at GIAC 2026 Shenzhen on Agent Engineering with @LangChain . The real work isn't prompt engineerin…
Summary
作者分享了在GIAC 2026深圳会议上关于Agent Engineering的演讲，强调构建可靠、有状态的代理的重要性。
View Cached Full Text
Cached at: 06/27/26, 06:00 PM
🎙️ Just wrapped my talk at GIAC 2026 Shenzhen on Agent Engineering with @LangChain .
The real work isn’t prompt engineering — it’s building agents that are reliable, stateful, and actually finish tasks.
Great energy from the 🇨🇳 Shenzhen crowd! Recap of the stack 👇 https://t.co/kq2A4v5lLd
Similar Articles
@LangChain: Agents are writing code, processing files, analyzing data, installing packages, and running multi-step workflows. To do…
LangChain is hosting a technical webinar on July 15 about building a secure execution environment for AI agents, covering security, isolation, and observability.
@LangChain: Giving agents the ability to write code makes them dramatically more capable. It also makes security a lot harder. At L…
LangChain's Deep Agents allow untrusted agent-written code to run securely using WebAssembly-based code interpreters, providing execution and capability isolation without a traditional sandbox.
@Kimberl9633: LangChain just released two major updates today, directly pushing agent reliability forward: a unified evaluation stack, and a code execution solution that doesn't require a full sandbox. First, evaluation. How to test long-running, stateful agents? The combination of Harbor + LangSmith provides…
LangChain has released a unified evaluation stack (Harbor + LangSmith) and an in-process code execution solution based on WASM+QuickJS, aimed at improving the evaluation reliability and execution safety of AI agents.
@rajeevchhajer: Key ideas I picked up at the @LangChain conference this week: "AI engineering is data science. Look at your data." — @s…
Key takeaways from the LangChain conference, including insights on AI engineering as data science, the priority of data strategy before agents, coding agents as the new substrate, and context plus memory as the new moat.
@LangChain: What do the agents of the future look like? A highlight from @hwchase17's Day 2 keynote.
LangChain announces on-demand interrupt capability for agents, highlighted in Harrison Chase's Day 2 keynote.