--- title: "Qiushi (@orangeclaw-stack) | Hashnode" description: "AI agents that remember, coordinate, and never sleep. Read the latest articles by Qiushi on Hashnode." url: "https://hashnode.com/@orangeclaw-stack" --- @orangeclaw-stack # Qiushi @orangeclaw-stack · New York, USA · Joined March 2026 AI agents that remember, coordinate, and never sleep. ### [Teaching an AI to Attend Your Teams Meeting: A Real-Time Voice Pipeline](https://claw-stack.hashnode.dev/teaching-an-ai-to-attend-your-teams-meeting-a-real-time-voice-pipeline?utm_source=hashnode&utm_medium=feed) Microsoft Teams has a lot of AI features now. Copilot can summarize meetings after the fact. It can generate action items. It can transcribe. What it cannot do — at least not in any form that worked for us — is let you summon an AI agent mid-meeting ... 12 min read · [claw-stack.hashnode.dev](https://claw-stack.hashnode.dev/teaching-an-ai-to-attend-your-teams-meeting-a-real-time-voice-pipeline?utm_source=hashnode&utm_medium=feed) [Mar 16](https://hashnode.com/posts/teaching-an-ai-to-attend-your-teams-meeting-a-real-time-voice-pipeline/69b789f242a134433ce26004) ### [From Code Completion to Code Team: How We Turned Claude Code into an Engineering Department](https://claw-stack.hashnode.dev/from-code-completion-to-code-team-how-we-turned-claude-code-into-an-engineering-department?utm_source=hashnode&utm_medium=feed) We use Claude Code every day. It's excellent. It handles complex refactors, writes tests, navigates large codebases, and catches bugs we'd miss. But after months of running it as part of an autonomous multi-agent system, we noticed something: Claude ... 7 min read · [claw-stack.hashnode.dev](https://claw-stack.hashnode.dev/from-code-completion-to-code-team-how-we-turned-claude-code-into-an-engineering-department?utm_source=hashnode&utm_medium=feed) [Mar 10](https://hashnode.com/posts/from-code-completion-to-code-team-how-we-turned-claude-code-into-an-engineering-department/69b049a38288a887ce10b2b6) ### [Memory System v2: Solving the Context Bloat Problem](https://claw-stack.hashnode.dev/memory-system-v2-solving-the-context-bloat-problem?utm_source=hashnode&utm_medium=feed) In our last post on building a persistent memory system, we described the MEMORY.md bloat problem: after six weeks, the file had grown to over 700 lines, and we fixed it by switching from inline content to pointer-based entries. The fix worked. MEMOR... 10 min read · [claw-stack.hashnode.dev](https://claw-stack.hashnode.dev/memory-system-v2-solving-the-context-bloat-problem?utm_source=hashnode&utm_medium=feed) [Mar 9](https://hashnode.com/posts/memory-system-v2-solving-the-context-bloat-problem/69af38cefff0617bc7751819) ### [Making iMessage Reliable with OpenClaw: 3 Problems and How We Fixed Them](https://claw-stack.hashnode.dev/making-imessage-reliable-with-openclaw-3-problems-and-how-we-fixed-them?utm_source=hashnode&utm_medium=feed) OpenClaw can use iMessage as a communication channel — you text your AI agent, it texts you back. Sounds simple, but running it 24/7 on a Mac mini revealed three reliability issues that took weeks to fully diagnose. Here's what went wrong and how we ... 6 min read · [claw-stack.hashnode.dev](https://claw-stack.hashnode.dev/making-imessage-reliable-with-openclaw-3-problems-and-how-we-fixed-them?utm_source=hashnode&utm_medium=feed) [Mar 9](https://hashnode.com/posts/making-imessage-reliable-with-openclaw-3-problems-and-how-we-fixed-them/69ae1a6edf0208e5a0d1081d) ### [Building a Tri-Modal Knowledge Engine for CTF Agents](https://claw-stack.hashnode.dev/building-a-tri-modal-knowledge-engine-for-ctf-agents?utm_source=hashnode&utm_medium=feed) When Librarian gets asked about tcache stashing, it needs to return something more useful than what a base Claude model knows. The model has a general understanding of heap exploitation — it can describe what tcache is, explain the concept of a stash... 8 min read · [claw-stack.hashnode.dev](https://claw-stack.hashnode.dev/building-a-tri-modal-knowledge-engine-for-ctf-agents?utm_source=hashnode&utm_medium=feed) [Mar 7](https://hashnode.com/posts/building-a-tri-modal-knowledge-engine-for-ctf-agents/69acbbb7a9dd883e9b88083f)