Agentic AI built for your operations.

The patterns behind these consulting engagements are documented openly on the blog. Use the sections below as a reading map — each cluster matches a phase of a typical engagement.

Cost discipline and model routing

For cost discipline, start with pre-agentic data fetching and the 19–62% token cut and the prompt-caching playbook that drops repeat-call cost 47%.

For model tiering before you touch routing code, the visual note stop paying frontier prices for classification is the map I hand to teams on day one.

Orchestration, tools, and memory

For orchestration, read why I'm replacing supervisor patterns with handoffs in multi-agent systems.

For registry hygiene that drives tool-selection accuracy in production, tool descriptions are prompts — fix the registry, not the agent.

The first audit on every engagement is the tool contract itself: your agents aren't broken, your tools are runs the three questions (atomic, honest on failure, typed) I check before touching a prompt.

For agent memory architecture — the layer most production agents are silently missing — the three paradigms of LLM memory (implicit, explicit, agentic) is the map we use in every engagement.

For the whole picture in one place, the anatomy of an AI agent is the four-box diagram (memory, tools, the loop, guardrails) I draw at the start of every engagement.

Observability and evals

Every engagement ships with the agent observability stack we deliver to every client and eval datasets that go beyond the happy path.

For a one-glance carousel summary of the tool-design rules, see the visual note on fixing wrong-tool calls.

Architecture choices and reference implementations

Before committing to an agent architecture at all, read AI agent vs agentic AI and what the distinction actually means when you ship one. That post is the framing I use to scope every engagement.

Once the architecture is settled, the next decision is execution shape: code agents vs skill agents and when to pick which walks through the keyboard-vs-toolbox tradeoff and the hybrid pattern most production systems become.

For the implementation layer inside both shapes, tool registry design for agentic AI is the audit recipe I run on every client codebase before changing anything else.

For a worked example of all these patterns stitched together, Recruiting Atelier is the runnable reference I built: ReAct loop, supervisor with planning, tool registry, MCP integration, RAG, guardrails, and observability.

For a planning-layer architecture at enterprise depth (12 stages, 17 guardrails, task harnessing, IDE prompts), see the Planner Agent reference spec.

May 2026: routing, MCP, and supply chain

For the routing-layer question after Google's Gemini 3.5 Flash launch, Gemini 3.5 Flash vs Sonnet 4.6 and the cost-per-completed-task maths is the migration recipe I am running at clients this week.

For enterprise MCP governance after Databricks shipped Unity AI Gateway, what changes for enterprise agents and the migration I would run for a Databricks shop covers the four-primitive build-vs-buy line.

On the protocol itself, the May 21 MCP spec release candidate takes the protocol stateless — the architecture shift I factor into every new MCP build.

After a poisoned VS Code extension harvested Claude Code credentials, hardening the agent supply chain (extensions, skills, MCP configs, and keys) is now the first review on any production engagement.

June 2026: billing, governance, and grounded retrieval

For the same-day Claude plumbing cutover (Agent SDK credit split plus Opus 4 / Sonnet 4 API retirement), the June 15 billing checklist I run before anything breaks at 2 a.m. covers auth paths, credit claims, and model ID greps.

For Cursor Auto-review and pre-push /review as an autonomy dial (not a security boundary), governing agent autonomy in 2026 is the adoption path I am rolling out on client repos this week.

For frontier routing after Fable 5 shipped and suspended in the same week, Claude Fable 5 for agent builders covers retention, safeguard fallbacks, and fallback IDs.

For Google's Agentic RAG preview with a Sufficient Context Agent, when iterative retrieval beats retrieve-then-pray is the decision table I use before re-architecting a knowledge base.

For content QA pipelines (GEO + SEO eval before publish), the visual note agentic AI content quality: 5 agents, one pipeline is the architecture I run on pages before they go live.

July 2026: MCP GA, swarm routing, sandbox evals, Claude Code Week 28–29

After MCP 2026-07-28 went live, the day-one field notes cover MRTR, header routing, handle triage, and the first-48-hour watch list, paired with the pre-GA cutover checklist.

For Cursor's planner/worker swarm economics (and how I map them onto production graphs), the swarm routing playbook sits next to the unified spend dashboard.

After Anthropic's July 30 cyber-eval disclosure, the agent eval sandbox checklist is the egress-proof and incident tempo I require before raising autonomy on networked agents.

For Claude Code dynamic workflows and the 14-step graph engineering roadmap (nodes, edges, diamonds, verifiers), graph engineering with Claude Code is the long-form, and the visual note is the one-page map I share in workshops.

For Claude Code's Desktop browser and upgraded /doctor (v2.1.202–v2.1.206), the Week 28 rollout checklist covers domain allowlists, /doctor confirmation culture, and auto-mode transcript protections.

For Week 29 live MCP artifacts (v2.1.207–v2.1.212), the live connector artifact checklist covers viewer-scoped OAuth, public link policy, and long-MCP auto-background.

After mid-July reporting on a Google/Microsoft/Salesforce/Snowflake/ServiceNow shared enterprise agent protocol framed against MCP, the MCP portability playbook is how I keep tool contracts and IdP policy above the wire while finishing the July 28 migration.

For the July 21 Gemini efficiency launch (3.6 Flash, 3.5 Flash-Lite; Pro still partner-only), the Flash routing checklist updates cost-per-completed-task lanes without betting the roadmap on Pro.

Handle migration and header hygiene still matter after GA: week-three handle migration and the header security checklist.

Topic pillars

For the full landscape, browse: Agentic AI, multi-agent systems, Model Context Protocol, AI observability, and enterprise AI automation.