Blog

Field notes on architecture, AI enablement, compliance systems, and product engineering.

Featured image for AI Coding Works Better When the Agent Has a Contract

AI Coding Works Better When the Agent Has a Contract

A practical field guide to structuring software work for AI coding agents using persistent documentation, task contracts, implementation plans, bounded slices, preflight audits, and verification.

Featured image for AI Coding Agents Are Great. Until Someone Asks You to Code Without One.

AI Coding Agents Are Great. Until Someone Asks You to Code Without One.

A practical look at how AI coding agents are changing developer workflows, which skills live coding interviews still measure well, and where modern hiring may be testing the wrong things.

Featured image for The Demo Went Well. Then Procurement Sent the Security Questionnaire.

The Demo Went Well. Then Procurement Sent the Security Questionnaire.

A practical SaaS compliance walkthrough showing how to distinguish a defensible Yes, a qualified answer, and an Unknown using assessment results and underlying evidence.

Featured image for Your AI Agent Is a Privileged User. Architect It Like One.

Your AI Agent Is a Privileged User. Architect It Like One.

A practical guide to securing AI agents in SaaS with least privilege, tenant isolation, deterministic authorization, approval gates, usage limits, cost controls, and auditability.

Featured image for The Missing Compliance Artifact Wasn't Another Policy

The Missing Compliance Artifact Wasn't Another Policy

A practical ISO 27001 readiness case study showing how existing evidence, supplier security gaps, operational proof, and reassessment fit together in a SaaS compliance workflow.

Featured image for Why .NET Is Still Relevant in 2026

Why .NET Is Still Relevant in 2026

A practical look at why .NET remains relevant in 2026 for backend systems, SaaS products, long-lived business software, maintainability, diagnostics, and AI-adjacent development.

Featured image for Why .NET Developers Don't Need to Switch to Python for AI

Why .NET Developers Don't Need to Switch to Python for AI

A practical guide for C# developers evaluating .NET vs Python for AI, covering Microsoft.Extensions.AI, Semantic Kernel, RAG, vector search, evaluations, tokenization, and local models.

Featured image for Clean Architecture, Vertical Slice, and Feature Modules: What Actually Helps in Real Projects

Clean Architecture, Vertical Slice, and Feature Modules: What Actually Helps in Real Projects

A practical look at three ideas teams often mix together, and what usually matters more than the diagram when a SaaS product has to keep changing.

Featured image for Why Early-Stage SaaS Usually Needs a Modular Monolith, Not Microservices

Why Early-Stage SaaS Usually Needs a Modular Monolith, Not Microservices

A practical comparison of modular monoliths and microservices for early-stage SaaS, focusing on delivery speed, operational complexity, boundaries, and when distribution actually becomes useful.

Featured image for Why I Still Like Owning the Deployment Box

Why I Still Like Owning the Deployment Box

A practical comparison of VPS hosting and managed cloud platforms for SaaS deployment, including cost, operational control, Docker-based hosting, vendor lock-in, and infrastructure trade-offs.

Featured image for Software That Survives Contact With Business Reality

Software That Survives Contact With Business Reality

Why this blog exists, and why useful software is less about impressive demos than about systems that keep working when real organizations depend on them.