All guides, page 8 of 21

Newest first. Search or pick a topic.

482 guides, newest first

• Out of Order HTML Streaming: Show the Page Shell First, Slow Parts LaterFrontend · A plain-English guide to out-of-order HTML streaming: named placeholders, the template element's for attribute, a Python streaming server, proxy buffering and fallbacks, shown on one running example of a travel booking page.19 min read · Updated 23 Oct 2026
• Prompt Injection Prevention: How to Defend AI AgentsAI and GenAI · A plain-English guide to prompt injection and how to defend an AI agent against it, shown on one email assistant from the first attack to a safer design and a test plan.28 min read · Updated 23 Oct 2026
• RAG Chunking Strategies: How to Split Documents for RetrievalAI and GenAI · A plain-English guide to splitting documents for RAG, from chunk size and overlap to structure-aware, semantic and parent-child chunking, shown on one running example of a farmer helpline.27 min read · Updated 23 Oct 2026
• System Prompt Best Practices: Guidance, Not EnforcementAI and GenAI · A plain-English guide to writing good system prompts, why models can still break them and which rules must move into code, shown on one running example of a savings app's chat helper.21 min read · Updated 23 Oct 2026
• A/B Testing LLM Applications: Metrics, Splits and Safe RolloutAI and GenAI · A plain-English guide to A/B testing LLM applications: offline checks, shadow tests, metrics, user splits, test length and a safe rollout, shown on one running example of two product description prompts.24 min read · Updated 23 Oct 2026
• Choosing an AI Agent Framework: Plain Loop, Graph-Based or Role-BasedAI and GenAI · A plain-English guide to choosing an AI agent framework: a plain loop, LangGraph, CrewAI, Pydantic AI or Microsoft Agent Framework, tested on one running example of a small tax-assistant team.21 min read · Updated 23 Oct 2026
• AI System Design Case Study: AI-Assisted Loan OriginationAI and GenAI · A plain-English guide to an AI system design case study: a loan system where AI reads documents and drafts a risk summary but a person decides, shown on one running example of a small-business lender.31 min read · Updated 23 Oct 2026
• AI SLA: How to Turn AI Trade-offs into Promises You Can KeepAI and GenAI · A plain-English guide to writing an SLA for an AI service: SLIs, SLOs, quality audits, error budgets, fallbacks and client reports, shown on one running example of an invoice extraction service.27 min read · Updated 23 Oct 2026
• AI Agent Authorization: Authentication vs Authorization GuideAI and GenAI · A plain-English guide to AI agent authorization: authentication vs authorization, OAuth scopes and tokens, service accounts, the confused deputy problem, token storage and MCP, shown on one running example of an invoicing assistant.28 min read · Updated 23 Oct 2026
• AI Solution Architecture: From Business Problem to DesignAI and GenAI · A plain-English guide to AI solution architecture: discovery questions, success metrics, the no-AI check, building blocks, a first diagram and a phased rollout, shown on one running example of a lunch-box delivery complaints desk.27 min read · Updated 23 Oct 2026
• How to Debug LLM Hallucination: Prompt, Context or Model?AI and GenAI · A plain-English guide to debugging LLM hallucination: reproduce the wrong answer, read the trace, prove the cause with one change and fix the right layer, shown on one insurance help bot.24 min read · Updated 23 Oct 2026
• Design Distributed Search: Inverted Indexes and ShardsSystem design · A plain-English guide to distributed search: how an inverted index works, and how shards, replicas and a coordinator spread search across machines, built on one running marketplace example.25 min read · Updated 23 Oct 2026
• AI Compliance Architecture: GDPR, HIPAA and FedRAMP ExplainedAI and GenAI · A plain-English guide to what GDPR, HIPAA, FedRAMP and the EU AI Act mean for how you build an AI system, shown on one running example of a clinical note-taking app.27 min read · Updated 23 Oct 2026
• LLM Model Selection Per Task: Tiering, Routing and CascadesAI and GenAI · A plain-English guide to choosing an LLM for each task: pass marks, small test sets, tiers, routers, cascades and fallbacks, shown on one running example of a grocery delivery app.25 min read · Updated 23 Oct 2026
• Human in the Loop AI: Designing Escalation and ApprovalsAI and GenAI · A plain-English guide to deciding which AI decisions need a person, when to escalate, how to design the reviewer screen and how to stop rubber-stamping, using one motor insurance claim from start to finish.25 min read · Updated 23 Oct 2026
• AI Agent Security: Least Privilege and Fixing Tool BloatAI and GenAI · A plain-English guide to applying least privilege to AI agents, from auditing a bloated tool list to narrow tools, scoped short-lived credentials, approvals and logs, shown on one HR assistant from start to finish.28 min read · Updated 23 Oct 2026
• LLM Latency Optimization: Accuracy vs Latency Trade-offsAI and GenAI · A plain-English guide to LLM accuracy vs latency: where the wait comes from, how to measure it at p50 and p95, how to cut it safely and when accuracy must win, shown on one running example of a driver voice assistant.27 min read · Updated 23 Oct 2026
• LLM Cost Optimization: Cut Your AI Bill Without Hurting QualityAI and GenAI · A plain-English guide to LLM cost optimization: how token pricing works, how to measure cost per feature and per user, and which fixes cut the bill, shown on one running example from a meeting-notes startup.30 min read · Updated 23 Oct 2026
• LLM Data Residency: Where Your AI Actually RunsAI and GenAI · A plain-English guide to LLM data residency: what data leaves your app, the four places a model can run, what to ask about retention and training, and DPDP and GDPR transfer basics, all shown on one running clinic example.26 min read · Updated 23 Oct 2026
• LLM Evaluation Metrics and Test Frameworks ExplainedAI and GenAI · A plain-English guide to testing an LLM app: golden test sets, the main evaluation metrics, LLM-as-judge, error margins and regression tests in CI, all shown on one running example of a pharmacy chat assistant.28 min read · Updated 23 Oct 2026
• LLM Feedback Loop Design: Turning User Feedback into FixesAI and GenAI · A plain-English guide to designing a feedback loop for an LLM product, from collecting thumbs, edits and overrides to safe fixes, shown on one running example of AI essay feedback.22 min read · Updated 23 Oct 2026
• LLM Guardrails Explained: Layered Checks Before and After the ModelAI and GenAI · A plain-English guide to LLM guardrails as layered defence: input checks, retrieval and tool limits, output checks, human review and testing, shown on one running example of a children's homework app.29 min read · Updated 23 Oct 2026
• LLM Observability: Catching Silent Regressions and Cost DriftAI and GenAI · A plain-English guide to LLM observability: what to log, how to trace multi-step calls, which alerts catch quality drops and rising cost, and how to find the cause, shown on one running example of a telecom chat assistant.25 min read · Updated 23 Oct 2026
• MCP vs API vs Agent-to-Agent: Which Integration to UseAI and GenAI · A plain-English guide to MCP vs API vs agent-to-agent protocols such as A2A: who decides each call, how they fit together and when to pick each, shown on one running example of a travel booking app.25 min read · Updated 23 Oct 2026