https://cirrusly.me Essays and articles on operational AI, distributed systems, and security. en-us https://cirrusly.me/From_SBOM_to_Decision_BOM.pdf https://cirrusly.me/From_SBOM_to_Decision_BOM.pdf Thu, 16 Jul 2026 12:00:00 GMT How inventories create false confidence in AI decisions — a complete decision record is not the same as sufficient evidence. Companion to Evidence Debt. https://cirrusly.me/Evidence_Debt.pdf https://cirrusly.me/Evidence_Debt.pdf Sun, 12 Jul 2026 12:00:00 GMT The system was designed to operate, not to prove what it did. https://cirrusly.me/OperationalAI_SAR.pdf https://cirrusly.me/OperationalAI_SAR.pdf Thu, 02 Jul 2026 12:00:00 GMT One system, decomposed three ways: by function, by placement, and by authority. https://cirrusly.me/Operational_AI_Distributed_Systems.pdf https://cirrusly.me/Operational_AI_Distributed_Systems.pdf Mon, 29 Jun 2026 12:00:00 GMT As agentic workloads scale, durable value moves from components to composition. https://www.linkedin.com/pulse/your-apis-actually-agentic-ready-stephen-alexander-azpie/ https://www.linkedin.com/pulse/your-apis-actually-agentic-ready-stephen-alexander-azpie/ Thu, 12 Mar 2026 12:00:00 GMT Why APIs must be redesigned for autonomous AI agents as primary consumers. https://www.linkedin.com/pulse/features-easy-validation-should-too-stephen-alexander-phpke/ https://www.linkedin.com/pulse/features-easy-validation-should-too-stephen-alexander-phpke/ Fri, 27 Feb 2026 12:00:00 GMT Building a smoke-test system to catch what unit tests miss in AI-generated code. https://www.linkedin.com/pulse/trust-verify-adversarial-review-ai-code-stephen-alexander-gch9e/ https://www.linkedin.com/pulse/trust-verify-adversarial-review-ai-code-stephen-alexander-gch9e/ Tue, 24 Feb 2026 12:00:00 GMT Adversarial AI code review surfacing real findings that passed traditional testing. https://www.linkedin.com/pulse/specs-vs-vibes-difference-between-building-hoping-stephen-alexander-fagoe/ https://www.linkedin.com/pulse/specs-vs-vibes-difference-between-building-hoping-stephen-alexander-fagoe/ Fri, 13 Feb 2026 12:00:00 GMT Spec-driven development with AI agents versus vibe coding. https://www.linkedin.com/pulse/building-ai-learning-hub-stephen-alexander-f1fye/ https://www.linkedin.com/pulse/building-ai-learning-hub-stephen-alexander-f1fye/ Tue, 10 Feb 2026 12:00:00 GMT Building a knowledge platform with Claude Code — why hooks, not prompts, are the guardrails. https://www.linkedin.com/pulse/til-what-i-learned-studying-agentic-ai-practice-stephen-alexander-er2ie/ https://www.linkedin.com/pulse/til-what-i-learned-studying-agentic-ai-practice-stephen-alexander-er2ie/ Mon, 26 Jan 2026 12:00:00 GMT 700+ controlled runs showing context, not just capability, determines agentic outcomes. https://www.linkedin.com/pulse/building-agent-workflow-observatory-study-ai-agents-stephen-alexander-ouqne/ https://www.linkedin.com/pulse/building-agent-workflow-observatory-study-ai-agents-stephen-alexander-ouqne/ Sun, 25 Jan 2026 12:00:00 GMT Observing how AI agents make decisions under controlled constraints. https://www.linkedin.com/pulse/lessons-from-human-ai-partnership-coding-beyond-lines-alexander-ldc1e/ https://www.linkedin.com/pulse/lessons-from-human-ai-partnership-coding-beyond-lines-alexander-ldc1e/ Tue, 28 Jan 2025 12:00:00 GMT Why effective human-AI partnerships depend on context, design, and treating AI as a teammate. https://www.linkedin.com/pulse/starting-end-mind-how-amazons-working-backwards-ai-stephen-alexander-pyoge/ https://www.linkedin.com/pulse/starting-end-mind-how-amazons-working-backwards-ai-stephen-alexander-pyoge/ Mon, 20 Jan 2025 12:00:00 GMT Applying Amazon's Working Backwards method to AI projects. https://d1.awsstatic.com/whitepapers/digital-transformation-and-it-modernization-for-elections-in-aws.pdf https://d1.awsstatic.com/whitepapers/digital-transformation-and-it-modernization-for-elections-in-aws.pdf Mon, 13 Jan 2025 12:00:00 GMT An AWS whitepaper on modernizing election systems for security and scale. https://www.linkedin.com/pulse/cloud-sayings-showdown-which-ai-model-comes-out-top-stephen-alexander-0vy5e/ https://www.linkedin.com/pulse/cloud-sayings-showdown-which-ai-model-comes-out-top-stephen-alexander-0vy5e/ Mon, 13 Jan 2025 12:00:00 GMT Comparing Claude and GPT models across response time, output, and user feedback. https://www.linkedin.com/pulse/from-idea-code-what-i-learned-using-ai-build-websites-alexander-rwnxe/ https://www.linkedin.com/pulse/from-idea-code-what-i-learned-using-ai-build-websites-alexander-rwnxe/ Tue, 17 Dec 2024 12:00:00 GMT Practical lessons from building with AI tools like Claude, Cursor, and v0. https://www.linkedin.com/pulse/ai-eating-world-culinary-journey-models-approaches-stephen-alexander-qbcse/ https://www.linkedin.com/pulse/ai-eating-world-culinary-journey-models-approaches-stephen-alexander-qbcse/ Mon, 09 Dec 2024 12:00:00 GMT Demystifying AI model types through restaurant analogies. https://www.linkedin.com/pulse/did-i-build-llm-gateway-curious-developers-journey-stephen-alexander-lubke/ https://www.linkedin.com/pulse/did-i-build-llm-gateway-curious-developers-journey-stephen-alexander-lubke/ Mon, 02 Dec 2024 12:00:00 GMT Accidentally building an LLM gateway while handling multiple foundation models. https://www.linkedin.com/pulse/my-journey-ai-security-what-every-cybersecurity-should-alexander-ganue/ https://www.linkedin.com/pulse/my-journey-ai-security-what-every-cybersecurity-should-alexander-ganue/ Mon, 18 Nov 2024 12:00:00 GMT Key AI security threats: adversarial ML, prompt injection, and AI-powered social engineering. https://docs.aws.amazon.com/whitepapers/latest/classic-intrusion-analysis-frameworks-for-aws-environments/classic-intrusion-analysis-frameworks-for-aws-environments.html https://docs.aws.amazon.com/whitepapers/latest/classic-intrusion-analysis-frameworks-for-aws-environments/classic-intrusion-analysis-frameworks-for-aws-environments.html Wed, 31 Mar 2021 12:00:00 GMT Adapting traditional intrusion analysis frameworks to the cloud.