← Back to the course home

⏮️ Before · why · "agentic" · the future

Every tool replaced something worse and points at something next. This page tells the agents story in four parts: what came before, why agents were born, what the word "agentic" actually means, and an honest look ahead.

⏮️ Before agents — the five eras

Each era kept the previous one's gift and fixed its gap.

⚙️ scripts & RPAwind-up toys1🗣️ chatbots'22-23: all advice2🧰 tool calling2023: first passes3🔁 agent loops'23-24: hype+crash4🔌 todaygated + MCP5📉 the AutoGPT lesson: loops without guardrails and evals ran in circles —the crash taught the field lessons 05 and 06each era kept the previous one's gift and fixed its gap

⚙️ Scripts & RPA (the long before): automation meant wind-up toys — exact recorded steps, brittle as glass. RPA bots clicked pixel coordinates and shattered when a button moved. Anything requiring judgment stayed human.

🗣️ Chatbots (2022–23): suddenly machines could REASON in language — brilliant advisors with no hands. "Here's how you'd book that trip" … but nothing got booked. All advice, zero outcomes (lesson 01's reciter).

🧰 Tool calling (2023): model APIs learned to emit structured function calls — the first hall passes 🎫. One call per turn: useful, but a human still drove the sequence.

🔁 Agent loops (2023–24): ReAct wired reasoning to tools in a LOOP — and the hype arrived: AutoGPT-style "fully autonomous" agents amazed on demos and then ran in circles, burned tokens, drifted off-goal. The crash was the curriculum: loops without guardrails (L05) and evals (L06) fail — publicly.

🔌 Today: production agents are narrower and humbler — scoped tools, write gates, budgets, logs — and MCP standardizes the tool belt so capabilities are shared parts, not private wiring. Boring-on-purpose is winning again (the k8s course's motto returns).

🐣 Why agents were born — six forces

No single invention — a convergence:

💰 The advice–outcome gap

  • advice is cheap; OUTCOMES are what work actually is
  • the whole economic prize sat one loop away from chatbots

🧠 Models crossed the planning bar

  • good-enough multi-step reasoning + reliable structured output
  • before that bar, loops just compounded nonsense faster

🧰 Tool-calling APIs

  • vendors made "model proposes a call" a first-class primitive
  • the harness pattern became easy to build (you read one in L07!)

🪑 Bigger desks

  • long contexts made scratchpads + tool results fit (L04)
  • without room to observe, the loop starves

🔌 Standard plugs (MCP)

  • tools became an ecosystem, not per-app wiring
  • N×M adapters → N+M (the MCP school's whole story)

📉 The hype crash itself

  • 2023's failures taught guardrails, evals, human gates
  • maturity, not magic, is what shipped in the end

🤔 What does "agentic" actually mean?

Agency = the capacity to pursue a goal by choosing and taking actions. Agentic = the adjective: how much of that capacity a system has. It's a DIAL, not a switch:

less agencymore agency0 · answer onlya chatbot reply01 · one tool callmodel fills arguments12 · chooses toolswhich + when23 · multi-step loopthink·act·observe34 · loops that WRITEgated by humans 🚧5 · long-horizon + budgets4“agentic” is a DIAL, not a switch — turn it only as far as your guardrails and evals have earnedgrammar: agent (the noun) · agency (the property: pursuing goals via chosen actions) · agentic (the adjective)

"Agentic workflow" = a pipeline where model-driven steps choose the path (vs a fixed script). "Agentic AI" in a product pitch = ask lesson 01's checklist: what's the goal? which tools? what loop? which guardrails? No answers = a chatbot in a trench coat 🕵️. And the dial has a price sticker: every notch right multiplies capability AND the need for L05's rules — turn it only as far as your evals prove you've earned.

🔮 The future — an honest outlook

Informed extrapolation, not prophecy — hold it loosely:

📈 near (now-ish)longer tasks · computer use ·real memory · evals as standard ·MCP tool ecosystems 🔌1🏢 nextteams of agents with foremen ·agent-ops as a job · humans shiftto reviewing & goal-setting 🚧2❓ open questionsreliability at long horizons ·injection & security · audit &accountability · workflow change3🧭 the safe bet: the LOOP stays; the leash lengthens as guardrails + evals maturewhatever the headlines say, it will still be think→act→observe with better tools and better brakes

📈 Near: longer-horizon tasks (hours, not minutes) as memory and self-verification improve · computer use (agents driving real UIs) maturing from demo to tool · evals and observability becoming as standard as tests are for code · MCP-style ecosystems making belts a marketplace.

🏢 Next: teams of agents with foreman patterns (L08) in real org charts · "agent ops" as a role (budgets, logs, evals, incident review — the k8s course's instincts, for loops) · humans shifting up the stack: setting goals, reviewing diffs, owning the signatures 🚧.

❓ Open questions the field hasn't earned answers to: reliability over very long horizons (compounding never sleeps — L06) · security against injection when agents read the open web · accountability and audit when loops act at scale · and how work itself reorganizes around drafts-flow-signatures-gate. Anyone selling certainty on these is selling. 😄

🧭 The safe bet: the loop stays. Whatever the headlines, it will still be think → act → observe — with better tools, longer leashes, and (if we're wise) better brakes. You already know the machine; the future is mostly its knobs.