Mottobits Fieldnotes · Adnan Rafiq & ChatGPT
AI meets the real codebase.
Co-written by Adnan Rafiq and ChatGPT. Practical fieldnotes on AI engineering, .NET performance, legacy QA, and dependable integrations—with sources, working methods, and explicit limits.
Subscribe by RSSAn AI-assisted .NET performance investigation, from trace to verified fix
Give an agent a bounded investigation, verify its hypotheses against runtime evidence, and prove the change under representative load.
6 MIN READ · PRACTICAL PLAYBOOKLegacy QABuild a regression safety net around the work that matters
Choose the first workflows to protect, build repeatable checks, and make every failure explainable before changing a legacy application.
5 MIN READ · PRACTICAL PLAYBOOKAI deliveryReview AI-generated tests before they become release evidence
A practical review contract for using AI to draft tests while preserving independent expectations, meaningful assertions, and human ownership.
4 MIN READ · PRACTICAL PLAYBOOK.NET modernizationMove one .NET capability at a time, with a route back
Scope an incremental ASP.NET migration around one capability, explicit compatibility checks, and a rollback decision that includes data.
5 MIN READ · PRACTICAL PLAYBOOKDynamics 365Design a Dynamics integration for the second attempt
Define ownership, retries, duplicates, and reconciliation before connecting business-critical data.
5 MIN READ · PRACTICAL PLAYBOOKInteractive assessmentCheck your release evidence
Ten questions, visible gaps, and a useful next step. No email gate or source upload.
FREE TOOL · RUNS IN YOUR BROWSERFrom the archive
Engineering briefs.
Source-backed notes from DotNetGuides. New briefs are kept in the publication repository and appear here when published.
Fix the background worker's scope before changing service lifetimes
A scoped-service error needs an ownership fix. Give your coding agent the registrations, work boundary, and checks that prove each job releases its dependencies.
2026-09-30 · ENGINEERING BRIEFPerformanceAn AI debugging brief that starts with evidence
Turn an exception into a focused investigation: observed facts, competing explanations, and the smallest test that separates them.
2026-09-29 · ENGINEERING BRIEFPerformanceAsk AI to review the SQL, not just the LINQ
A tidy LINQ expression can still move too much data. Build the review around the query, its plan, and its actual workload.
2026-09-29 · ENGINEERING BRIEFC#Give your AI a contract, not just a JSON sample
Generate the first draft of your C# models. Then make the assumptions explicit before an AI turns them into application code.
2026-09-29 · ENGINEERING BRIEFModernizationGive your coding agent one migration slice
A useful .NET migration task has a boundary, a behavior to preserve, acceptance tests, and a way back.
2026-09-29 · ENGINEERING BRIEFStart with a real problem
Want this thinking applied to your software?
Bring one slow workflow, one risky release, or one difficult integration. We’ll find a sensible first step.