My Portfolio https://www.raymondyxu.com This is my portfolio RSS feed Why Modern Realty Failed https://www.raymondyxu.com/blog/whyModernRealtyFailed With 3 million realtors in the USA, the IRL element of the CAC was insurmountable. Sun, 16 Aug 2026 00:00:00 GMT The long tail gets longer and fatter https://www.raymondyxu.com/blog/long-tail-gets-longer-and-fatter As agents search deeper, the long tail of online products and services gets evaluated more often—and purchasing spreads across more suppliers. Thu, 06 Aug 2026 00:00:00 GMT Label your AI writing as AI writing https://www.raymondyxu.com/blog/labelYourAIWritingAsAIWriting If you have a blog on your website, just fucking tell us which part is AI written. Tue, 04 Aug 2026 00:00:00 GMT Don't start a startup with the goal of getting acquired. https://www.raymondyxu.com/blog/dontStartAStartupToGetAcquired If your goal is to cash out at $5 million as quickly and reliably as possible, becoming a founder is a very strange way to do it. Mon, 27 Jul 2026 00:00:00 GMT What Is the Actual Point of Agentic Commerce? https://www.raymondyxu.com/blog/actual-point-of-agentic-commerce What agentic commerce is actually good for: saving buyers time, lowering sellers' acquisition costs, and making complex B2B purchases easier. Thu, 16 Jul 2026 00:00:00 GMT How does AI affect software company moats? https://www.raymondyxu.com/blog/howDoesAIAffectSoftwareCompanyMoats A thought experiment on which software company moats weaken or strengthen as coding costs fall and software users become agents. Tue, 16 Jun 2026 00:00:00 GMT Giving LLMs Psychedelics https://www.raymondyxu.com/blog/givingLLMsPsychedelics What if we want high-variance, creative LLMs instead of consistent ones? Exploring psychedelic-inspired inference time manipulations. Wed, 01 Apr 2026 00:00:00 GMT What if your agent owned its own repo? https://www.raymondyxu.com/blog/what-if-your-agent-owned-its-own-repo Agent-owned repos make vertical AI agents more consistent and deterministic by reusing verified code with execution behavior and reputation. Fri, 20 Feb 2026 00:00:00 GMT What's the Future of Agent-to-Agent Coding Collaboration Without Humans? https://www.raymondyxu.com/blog/agent-owned-repo Imagining a MoltBook-inspired future where agents can help each other write code without human intervention. Agents that work for different entities. Thu, 05 Feb 2026 00:00:00 GMT Claw-Con SF 2026 https://www.raymondyxu.com/blog/claw-con 775 photos from OpenClaw’s first AI/tech meetup in San Francisco, with AI-generated lobster captions by Claude Opus 4.6. Thu, 05 Feb 2026 00:00:00 GMT Semantic Code Caching: A Better Way to Do Long-Running LLM Agents https://www.raymondyxu.com/blog/semantic-code-caching Companies building long-running LLM agents hit a wall: context degrades quality, and debugging 200k token runs is brutal. The fix is semantic code caching. Mon, 02 Feb 2026 00:00:00 GMT Introducing Raysurfer https://www.raymondyxu.com/blog/introducing-raysurfer Announcing Raysurfer: AI maintained skills for long-running LLM agents. Re-use verified code from prior runs for many Xs of speedup and consistency improvements. Tue, 20 Jan 2026 00:00:00 GMT Code Mode Agents: Let the LLM write code, not call tools https://www.raymondyxu.com/blog/claudeCodeMode A library for using claude code to make its own workflow, live. Tue, 07 Oct 2025 00:00:00 GMT How Much Should You Tell Your AI Agent? https://www.raymondyxu.com/blog/howMuchAI How to guide your Long Running LLM agent with the correct amount of specificity. Tue, 30 Sep 2025 00:00:00 GMT