What is agentic AI in one sentence?

Software that picks tools and takes actions toward a goal — not just generates text — and recovers when a step fails.

Do I need an agent or is a script enough?

If the steps and their order are known in advance, write a script. Reach for an agent only when at least one step requires reasoning over the prior step's output.

What is the biggest mistake teams make shipping agents?

Skipping evaluation datasets and exit conditions. Most "the agent does not work" reports are missing one or both of those — the agent loops forever or fails silently because nothing scored its output.

Which model should I start with?

For most agent steps, start with Sonnet for reasoning and Haiku for routing. Move to Opus only when accuracy on a specific step justifies the cost.

Topic Pillar

Agentic AI.Systems that decide, act, and recover — built to ship.

Agentic AI is the next layer above generative AI: systems that pick tools, take actions, and recover from failure without a human in the loop on every step. This hub gathers the architecture patterns, cost-control techniques, and production lessons that make agents work outside the demo.

● 35 cluster pages· 16 posts· 3 notes· 14 updates· 2 events

What "agentic" actually means

An agent is not a chatbot with extra prompts. It is a system that picks a tool, takes an action against the real world, observes the result, and decides what to do next — with exit conditions, retries, and a budget. The difference between a prototype and a production agent is almost entirely in the boring parts: scope, observability, evaluation, and tool design.

When agents are the right tool

Use an agent when the work is multi-step, the steps depend on the previous outputs, and at least one step needs reasoning that scripts cannot encode. Skip agents for deterministic ETL, single-call classification, and any pipeline that runs the same five steps every time — those are scripts, and scripts are cheaper and more reliable.

The patterns that actually work in production

Pre-agentic data fetching, supervisor-vs-handoff orchestration, descriptive tool names, "when to use" descriptions on every tool, exit conditions on every loop, prompt caching as a first-class metric, evaluation datasets that go beyond the happy path, observability per step. These are the patterns we drill in training and ship in consulting.

16 blog posts

Deep dives on Agentic AI

Tool Design

Tool descriptions are prompts. Fix the registry, not the agent.

When an agent picks the wrong tool, the registry is broken — not the agent. Three rules I now apply before debugging anything in a multi-tool system: precise names, "when to use" triggers, and a curated load list. Anthropic's new tool-selection telemetry finally puts numbers on what changes accuracy.

May 13, 20265 min

Read the post Production

The cheapest LLM call is the one you do not make — GitHub's 19-62% token cut, decoded

GitHub published an instrumented analysis of their agentic CI workflows and reported 19-62% token-cost reductions. The savings are the headline. The technique — pre-agentic data fetching and tool-registry hygiene — is the story most teams will miss.

May 11, 20265 min

Read the post Architecture

Claude Opus 4.7's 1M context: when to RAG and when to just stuff it

A million tokens reliably is real now, but it does not retire RAG — it changes the calculus. Cost, latency, recency, and the prompt-cache angle nobody is talking about.

May 6, 20266 min

Read the post MCP

MCP 1.0 is here. What changes for the servers you already wrote

The protocol stabilised. Most working servers will keep working. Three places the new spec actually requires changes — auth profile, server registry, streaming-response semantics — with diffs from a real migration.

May 1, 20265 min

Read the post Multi-Agent

Why I am replacing supervisor patterns with handoffs

Supervisors looked clean on paper and shipped slow in production. Handoffs read messier in the code but recover better when an agent loses the plot. Two real systems and where supervisors still earn their keep.

Apr 26, 20266 min

Read the post Production

Prompt caching is not optional anymore — measuring a 47% cost drop

A walkthrough from a client engagement: identifying stable prefixes, restructuring the system prompt for cacheability, and the telemetry that proved caching was actually working.

Apr 19, 20264 min

Read the post Tool Design

Tool descriptions are prompts. Stop treating them like docstrings

A docstring tells a developer what a function does. A tool description tells a model when to call it. Different audience, different writing. Six concrete edits that lifted tool-call accuracy.

Apr 8, 20264 min

Read the post Production

The agent observability stack we ship to every client

Traces, spans, evals, cost-per-completed-task, and the one dashboard panel that catches 80% of regressions. Vendor-agnostic — covers Langfuse, Honeycomb, and rolling your own.

Mar 28, 20267 min

Read the post Architecture

Three patterns I broke in 2025 — and what I do instead now

Self-correction loops without budgets, single-agent solutions to multi-domain problems, and using JSON mode to force structure I should have built into the schema. An honest review.

Mar 14, 20268 min

Read the post Multi-Agent

Haiku 4.5 made our router 5x cheaper. The trade-off matters

Replacing Sonnet with Haiku in the dispatcher role cut our orchestration cost dramatically. It also cost us in two specific places I did not predict.

Feb 22, 20265 min

Read the post MCP

Why every team's first MCP server should be "list-files"

Smallest useful server. Hardest one to mess up. Teaches the protocol without distracting domain logic. The 60-line server we hand to teams during training.

Feb 4, 20263 min

Read the post Production

Eval datasets: stop testing your agents on the happy path

If your eval set is the demos you showed the client, you are testing the wrong thing. How we build evals from production failures and the minimum viable suite to ship.

Jan 19, 20266 min

Read the post Prompt Engineering

I was wrong about JSON mode. Here is what changed my mind

For two years I told teams to avoid forced JSON outputs and use structured tool calls. That was right then and partially wrong now — schema enforcement got better, latency penalties got smaller.

Dec 12, 20254 min

Read the post Architecture

Why your agent keeps failing after 3 steps

The exit condition problem nobody talks about. Most agents are built for the happy path — where every tool call succeeds and the task completes cleanly. Real production agents are different.

Nov 8, 20254 min

Read the post Tool Design

The one rule for designing agent tools that actually work

One tool, one purpose. Every tool that does two things will fail you on the third call. I have watched this pattern fail in every team I have trained — and the fix is the same refactor.

Oct 17, 20253 min

Read the post Architecture

RAG vs CAG: how to actually decide

A decision framework from real implementations. RAG retrieves. CAG stores in cache. Knowing which to use — and when to combine both — determines whether your agent finds the right answer at the right cost.

Sep 21, 20255 min

Read the post

3 carousel notes

Visual breakdowns on Agentic AI

Tool Design

Your agent called the wrong tool.

Fix the description. Not the agent.

Read the note

Tooling

Copilot vs Claude Code

One is fast. One is deep. Know which to use.

Read the note

Tooling

You merged it. Then it failed.

Modern pipelines review code before merge — automated review, security scan, and test generation on every PR.

Read the note

14 ship-news updates

Latest in Agentic AI

Claude

Anthropic ships tool-use telemetry — every selection is scored and logged at the model boundary

May 13, 2026 · via Anthropic

Tools

Claude Code adds parallel sub-agent execution — multi-file refactors land in a single turn

May 13, 2026 · via Anthropic

MCP

MCP remote-server registry crosses 500 listed servers — a curated production-ready tier emerges

May 12, 2026 · via modelcontextprotocol.io

Architecture

GitHub cuts agentic CI workflow costs 19-62% by pruning tools and moving data-fetch outside the LLM loop

May 11, 2026 · via GitHub Engineering Blog

Claude

Claude Opus 4.7 ships with 1M-token context window in production

May 7, 2026 · via Anthropic

Tools

Claude Code adds project memory — persistent context that survives across CLI sessions

May 5, 2026 · via Anthropic

MCP

MCP 1.0 ratified — official SDKs in Python, TypeScript, Go, Rust, Java, .NET

May 2, 2026 · via modelcontextprotocol.io

Architecture

Anthropic publishes "Effective Tool Design" — official guidance for production agents

Apr 28, 2026 · via Anthropic

Claude

Sonnet 4.6 update: cheaper tokens, sharper tool calls, fewer retry loops

Apr 24, 2026 · via Anthropic

See the full Agentic AI update feed

2 talks

Speaking on Agentic AI

11th DOT NET DAY

60 Minutes to Build Your First AI Agent

Saturday, 9 May 2026Ahmedabad

View recap & resources

The Agentic Operating System

How IT Companies Compete When AI Agents Work Cheaper Than Developers

Saturday, 2 May 2026Ahmedabad

View recap & resources

Enterprise solutions

How Agentic AI ships in our engagements

The pages below are the buyer-focused, conversion-grade versions of this topic — deliverables, methodology, ROI, security considerations, and CTAs to scope a real engagement.

Solution