Cloudflare Security Audit Skill: From Findings to Evidence
Explore Cloudflare’s security audit skill, its evidence-led workflow and the checks a software team should require before acting on AI findings.
AI Engineering 4 min read
AI Engineering
Retrieval, evaluation, routing, streaming and validation for LLM features that run in production, written from operating Conviro.
16 posts
Claudex Loop: Independent AI Review Before and After Coding
Learn how Claudex Loop separates planning, implementation and independent inspection, and when a cross-model review is worth the extra work.
AI Engineering 4 min read
WebMCP: Make Website Actions Understandable to AI Agents
Understand WebMCP’s browser-based tools, current experimental requirements and a practical pilot for helping agents use your website reliably.
AI Engineering 4 min read
MCP in 2026: What Stateless HTTP Changes for Integrations
Review MCP’s 2026 transport changes and build a migration plan that checks client compatibility, application state and recoverable business operations.
AI Engineering 4 min read
MCP Apps: When an AI Conversation Needs an Interface
Explore MCP Apps for interactive business workflows, from choosing records to reviewing actions, and learn what a useful first implementation needs.
AI Engineering 4 min read
Building and Operating Conviro: The Work Around the Model
Engineering lessons from building Conviro: tenant boundaries, retrieval quality, provider routing, tool permissions, and operating responsibility.
AI Engineering 4 min read
LLM Evals Should Exercise the Path You Actually Ship
Evaluate routing, retrieval, fallback, validation, and handoff together, then use diagnostic checks to explain where failures originate.
AI Engineering 3 min read
Velgrina AI Remodel: Separating Photo Understanding from Image Generation
Inside Velgrina’s AI Remodel feature: kitchen-photo understanding, image generation, daily limits and an administrative provider switch.
AI Engineering 3 min read
Velgrina: Building AI Product Help into an Online Store
A project note on Velgrina’s streaming AI product-help chat and the difference between a fluent answer and a useful shopping interaction.
AI Engineering 3 min read
LLM Budgets Must Survive Concurrent Requests
Turn a usage dashboard into an enforceable spending policy with atomic reservations, bounded calls, reconciliation, and tenant-aware accounting.
AI Engineering 3 min read
ConvergePilot: Turning an AI Audit into a Usable SaaS Product
What building ConvergePilot taught me about AI audit scope, subscription access, model selection and useful recommendations.
AI Engineering 3 min read
Multi-Provider LLM Routing Is a Policy Problem
A useful model router combines capabilities, permissions, region requirements, budgets, and failure behavior behind an explicit contract.
AI Engineering 3 min read