How to plan a RAG pilot for business knowledge
A practical guide to choosing documents, designing retrieval and measuring a grounded AI assistant for your business.
Read the noteInsights
Practical, source-backed guidance across AI engineering, software development, architecture, and DevOps. Each note includes its review date and primary sources.
A practical guide to choosing documents, designing retrieval and measuring a grounded AI assistant for your business.
Read the noteHow to scope, build and hand over a useful web application without losing sight of its users or operating needs.
Read the noteReview delivery pipelines, cloud reliability, security and spending with a prioritised plan instead of a tool-shopping list.
Read the noteCompare four ecommerce models and the practical decisions behind a storefront, supplier integrations and customer journey.
Read the noteA practical guide to defining capacity, ownership and deliverables when bringing senior engineering support into your team.
Read the noteWhat recent OpenAI and Anthropic model releases mean for business applications, and how to choose using your own evaluation data.
Read the noteRecent computer-use advances make new workflows possible. Here is how to design a bounded pilot with approval, audit history and recovery.
Read the noteA practical guide to model routing, prompt caching and evaluation, prompted by October 2026 API updates.
Read the noteA practical look at how we use OpenAI Codex to turn an idea into inspectable software, with human direction, tests and clear release boundaries.
Read the noteA practical checklist for teams reviewing disclosures, generated content, system documentation, and ownership after the EU transparency rules took effect.
Read the noteHow to turn Govern, Map, Measure, and Manage into concrete product, engineering, and operational work.
Read the noteAn application-level security review covering authority, data flow, integrations, and output handling using current OWASP guidance.
Read the noteWhat the DORA 2025 research means for engineering leaders adopting AI-assisted software development.
Read the noteHow teams can connect the current OWASP application risks to architecture, implementation, testing, and operations.
Read the noteA concise ADR structure for recording AI system boundaries, evaluation choices, fallbacks, and operational responsibilities.
Read the noteHow to make escalation and review usable, timely, and accountable instead of adding a ceremonial approval step.
Read the noteA practical way to standardise names and attributes before dashboards and alerts become difficult to trust.
Read the noteEngineering practices that make third-party software, build paths, and supplier responsibilities easier to understand and govern.
Read the noteA release review covering user value, data, evaluation, security, fallback, observability, and operational ownership.
Read the note