Jarvis Osorio's LinkedIn articles on AI agents: a reading guide
By Jarvis Nelson Osorio · November 16, 2026
Jarvis Osorio (Jarvis Nelson Osorio) writes regularly on LinkedIn about AI agents, the infrastructure around them and what happens when they meet real institutions. The complete, dated list lives at jarvisosorio.com/articles/. This post is a reading guide: where to start depending on what you care about.

If you build agents: start with the open-source "why" posts Two articles explain the reasoning behind his open-source tools:
- "Ambient Memory: Why I Built a Persistent State Layer for Autonomous Agents" (November 12, 2026) — the case for a local, encrypted memory layer that agents can query over MCP.
- "Agent Session Scrubber: Why I Open-Sourced a Sanitization & Pruning Engine for AI Coding Runtimes" (November 14, 2026) — why agent transcripts are a secrets problem, and why the tool defaults to dry-run.
Read them alongside the project pages on jarvisosorio.com and the repositories on GitHub.
If you care about agent safety and oversight A cluster of recent pieces argues that autonomous systems need structure, not just better prompts:
- "The AI Judge Paradigm: Why Multi-Agent Swarms Need Deterministic Oversight" (November 14, 2026)
- "The Banking Sector's Agent Nightmare: Why Financial AI Needs Hard Guardrails" (November 13, 2026)
- "Microsoft's Stark Warning: AI Agents Can't Be Trusted—Here Is the Architectural Fix" (November 12, 2026)
- "Why Is Your AI Agent Looping? The Root Cause, Economic Cost, and Architectural Fixes" (October 29, 2026)
- "The Rise of Agent Whales: How Autonomous Swarms and AI Bots Are Rewriting Market Manipulation" (October 28, 2026)
The common theme is the same one that runs through his code: constrain what agents can do, verify what they did, and make failure visible.

If you track models and AI infrastructure Jarvis Osorio also writes shorter analyses of new models and platforms, including:
- "Inside the Model's Mind: How Anthropic's Mechanistic Interpretability Is Making AI 'Thoughts' Visible" (October 16, 2026)
- "JetBrains Mellum2.1 Thinking: Why Local, High-Speed AI Agents Are the Future of Software Engineering" (October 15, 2026)
- "The Rise of System One Models: Why Celeris-1 Decision Is Redefining Agentic Architecture" (October 14, 2026)
- "OpenAI Dots: The Arrival of Always-On, Persistent AI Agents" (October 1, 2026)
- "Beyond Single Models: OpenRouter 'Fusion' is Changing the AI Architecture" (June 18, 2026)
- "OpenRouter: The API Revolution You Might Be Missing" (February 19, 2026)
If you want the human side Not everything is about architecture. "Look Up When Someone Enters the Room" (October 12, 2026) and "The Only Group Chat I Want to Be in Is with My AI Agents" (October 21, 2026) are more personal takes on attention and on living with AI tools. "Long-Term Return on Human Capital: How AI Systems Are Shaping the Global Workforce" (September 26, 2026) looks at work itself.
If you are curious about the website The newest entry, "Engineering the Infinite Plane: The Tech Stack and Spatial Scrolling Architecture behind my website JarvisOsorio.com" (November 15, 2026), walks through how his 3D homepage is built.
