Agentic engineering · Research

Clem
Ziroli III
— Agentic engineer

I design and build autonomous AI agents, and research what comes next for agentic engineering: longer-horizon agents, multi-agent orchestration, and the tooling and evals that make them dependable in production.

Fig. 01 — Agent topology Running
Diagram of a multi-agent system A planner delegates to three agents. The agents call tools, shared memory, and evaluations, and evaluation results feed back to the planner. PLANNER AGENT · A AGENT · B AGENT · C TOOLS MEMORY EVALS T0 · DELEGATE T1 · ACT T2 · VERIFY EVAL LOOP ↺
Plan Act Verify Repeat
Name
Clem Ziroli III
Role
Agentic engineer
Based
United States
Focus
Agentic engineering

01Research

Agentic engineering research

Autonomous AI agents

Researching the next level of agentic engineering.

Agentic engineering is the craft of turning large language models into autonomous AI agents — systems that plan, reason, use tools, and act. The work here is about pushing those agents further: more capable, more reliable, and more autonomous.

  • 01 Autonomous agents

    Autonomous AI agents

    Agents that plan, reason, and act on long-horizon tasks — with minimal human hand-holding.

  • 02 Orchestration

    Multi-agent orchestration

    Routing, delegation, and shared memory so fleets of specialized agents collaborate without falling over.

  • 03 Tooling & evals

    Agentic tooling & evaluations

    The tool integrations, guardrails, and evals that make agentic systems dependable in production.

02Releases

Releases

Free · No sign-up

In development

Coming soon.

I'm building a set of small, free web tools — including agentic ones. They'll live at CoolWebTools.net: simple, no sign-up, yours to use. Take a look, and check back as new ones land.

Destination
CoolWebTools.net
Price
Free
Accounts
None
Visit CoolWebTools.net

03About

About

Hi there

I'm Clem Ziroli III, an agentic engineer based in the United States.

I research and build autonomous AI agents: agent architecture, multi-agent orchestration, tool use, and evaluation. My focus is the next level of agentic engineering — agents that ship dependable, real-world work.

The free tools come out at CoolWebTools.net. No accounts, no upsells, no tracking — take whatever's useful.

More soon. Feel free to say hi.

04FAQ

Agentic engineering FAQ

Read me

Q1 What is agentic engineering?

Agentic engineering is the discipline of designing, building, and operating autonomous AI agents — software that uses large language models to plan, make decisions, call tools, and complete real tasks with little human intervention. It spans agent architecture, orchestration, memory, tool use, evaluation, and reliability.

Q2 What does an agentic engineer do?

An agentic engineer builds AI agents and the systems around them: context and prompt engineering, tool integration, multi-agent orchestration, memory, guardrails, and evaluations — turning raw language models into dependable autonomous systems.

Q3 What is the next level of agentic engineering?

The next level is agents that handle longer-horizon work, self-correct, and stay reliable in production — stronger planning, rigorous evaluation, and multi-agent collaboration that ships real results safely. That is the research focus here.