Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: agentic-ai
Cutting Agent Token Costs: Caching, Routing and Context Hygiene
Shrink LLM agent bills with prompt caching and semantic caching, lazy-loaded tools, model routing and cascades, subagents, and plain old context hygiene.
16 min readAgentic Analytics Is More Than Text-to-SQL: Lessons From Anthropic
What Anthropic's self-service analytics post says about semantic layers, skills, evals and data foundations, and why text-to-SQL alone gives a false sense of precision.
17 min readAI Architecture for Production Agents: Failure Modes and Control Layers
A layer-by-layer architecture guide for production agent systems: failure modes, guardrails, evals, observability, cost control, zero trust and governance.
12 min readA Sandbox for Claude Code: Isolated Agent Environments on a Laptop
How a local Kubernetes sandbox gives coding agents their own machine, scoped credentials, an egress allow-list and a fleet view, with build-versus-buy tradeoffs.
12 min readLoops vs. Graphs: Two Ways to Architect AI Agents
Loop engineering vs graph engineering for AI agents: how each architecture works, when to pick one over the other, and a checklist to use before production.
12 min readSafe Tool Discovery for Enterprise AI Agents: Registry, Gateway and Policy
A design for financial-services agents: a Tool Registry, filtered discovery, a deterministic Tool Gateway, intent-level authorization and durable side effects.
7 min readFrom Evals to Self-Improving Agents: Five Eras of Agent Evaluation
Agent evaluation now means auditing traces. See how deterministic checks, LLM and agent judges, diagnosis and a meta-harness build a safe improvement loop.
7 min readGenerative AI vs Agentic AI: What Really Sets Them Apart
Generative AI answers prompts while agentic AI pursues goals across steps. Learn how trust, risk and value differ and how to tell a real agent from marketing.
6 min readAgentic AI Testing for SaaS: How Goal-Driven Agents Replace Brittle Scripts
Why SaaS test suites keep breaking, how goal-driven AI agents explore an app and adapt, where they fall short, and how to run a measured agentic testing pilot.
6 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.