tecnovy AI Agentic AI Engineering

By the end of these two days an agent you built yourself is running: from a simple ReAct loop to a system that calls external tools, manages state and works in a cloud sandbox. Along the way you decide when an agent is worth building at all, how to secure it with guardrails and logging, and how to measure whether it actually works. The code is yours at the end, along with the architecture decisions behind it.

Attendance certificate4.8/5 on ProvenExpert

What this module delivers.

For developers and architects who want to run agents in production. You leave the two days with your own code, an architecture decision that holds up, and criteria against over-engineering. Your team can judge which tasks are worth automating agentically.

How the training runs

Method: Theory alternating with labs on an agent of your own
Basis: tecnovy curriculum, no external certification standard
Outcome: tecnovy certificate of participation, no external examination

Dates & booking

Choose a date that fits

No public dates are scheduled yet.

New dates are published regularly. Join the waitlist or plan an in-house course for your team.

Request in-house training

Fit

Who this module is designed for

Typical roles

You build applications on language models and are hitting the limits of simple prompts.
You design architectures and have to decide whether an agent is the right solution.
You are responsible for operating AI services safely, logging and guardrails included.

Prerequisites

No prerequisites

You can start right away. What you need: foundations in software development and Python, your own laptop with a working Python environment. Helpful: experience with AI technologies or cloud platforms.

Consider instead AI Practice and Management Two days of AI orientation with no coding, aimed at deciding rather than building.

Curriculum

Agentic AI Engineering Course in Detail

Seven blocks across two days, from the foundations of agentic systems to multi-agent patterns. Theory alternates with labs, and the agent you build grows from block to block.

01Agentic AI foundations

The opening block separates agents cleanly from what came before.

How they differ from chatbots and classic RAG applications
Planning, deciding and acting as the three building blocks
State, memory and tool access in outline

02Business cases and decision criteria

Here you decide whether a task deserves an agent at all.

Classes of task worth doing agentically, and those that are not
Cost, latency and error tolerance as selection criteria
When a plain workflow is the better solution

03Reasoning and planning patterns

This part introduces the patterns everything else builds on.

ReAct as the basic loop of reasoning and acting
Tool calling and structured output
Multi-step reasoning and task decomposition

04Architecture and orchestration

Now the pattern becomes a system with interfaces.

Single-agent and multi-agent architectures compared
Orchestrating workflows with cloud and open-source services
Connecting external tools, APIs and data sources

05Hands-on implementation

The longest lab block: your agent gains tools and memory.

Building up step by step from a simple agent to a tool user
Implementing memory and state management
Running it in a sandboxed cloud environment

06Evaluation, monitoring and optimisation

This part settles how you know the agent is any good.

Metrics for agentic systems beyond answer quality
Monitoring tool calls, cost and latency
Systematic tuning instead of guessing at prompts

07Security, governance and advanced patterns

The closing block makes the agent fit to operate.

Guardrails, logging and red teaming against misuse
Agentic RAG and human-in-the-loop as control patterns
Multi-agent orchestration and multimodal agents

Outcome

What you will be able to do afterwards

01
You decide with reasons which tasks should be solved agentically and which should not.
02
You apply agent patterns such as ReAct, tool calling and multi-step reasoning.
03
You design architectures for single-agent and multi-agent systems.
04
You build a working agent in Python and connect external tools and APIs to it.
05
You evaluate an agentic system and tune it against latency and cost.
06
You secure agents with guardrails, logging and red teaming.
07
You apply agentic RAG, human-in-the-loop and multimodal agents where they hold up.

tecnovy Teilnahmezertifikat

Confirms your attendance and the curriculum covered in the training. Issued by tecnovy, shareable on LinkedIn.

Open the Certificate Showroom tecnovy →

Why tecnovy

What you get on top with us

No Slideshow, Hands-On!

Promised: no PowerPoint marathon. We work in groups, tie theory to practice, and you get real project examples from our experienced trainers plus the exchange with like-minded people.

Certificate Showroom

Get your certificate of participation and, if you have one, add your exam certificate from E-Learning. Fully automated, beautifully designed. Just for you, only at tecnovy.

Flexible Date Change

If you are not able to attend the course, you can rebook your training free of charge up to one week before the start of the training.

Attend Twice, Pay Once

You are welcome to attend the training online again within a year as a refresher.

Learn from Experts

We always guarantee you the use of didactically and methodically first-class qualified trainers who draw their knowledge from training experience as well as professional practical and project experience.

FAQs

Frequently asked questions

01Is there a certification examination at the end?

No, and for a substantive reason: there is as yet no established international certification body for agentic AI. The tecnovy training relies on demonstrable practice instead and ends with a certificate of participation.

02Do I build a working agent during the training?

Yes. You start from a simple ReAct loop and extend it step by step with tool calls, memory and a sandbox environment. The code is yours and can be reused in your own projects.

03Which vendor does the course tie me to?

None. The labs run on Python, widely used open-source frameworks and cloud services on Azure. The patterns and decision criteria apply equally to other LLM providers and clouds.

04What prior knowledge do I actually need?

Foundations in software development and Python are required, because you write code yourself in the labs. Experience with AI or cloud platforms helps but is not essential.

05Are multi-agent systems covered?

Yes. The advanced part covers multi-agent orchestration, agentic RAG, human-in-the-loop and multimodal agents, each alongside the question of when a single agent remains the better choice.

06How does agentic AI differ from RAG or a chatbot?

An agentic system runs multiple steps, calls tools, accesses state and adapts its approach to intermediate results. A chatbot answers, a RAG application retrieves and answers; an agent acts.

07Can the training be tailored to our stack?

Yes. tecnovy tailors in-house deliveries to your technology stack and use cases, so that the labs build on your own systems.

What does your training at tecnovy look like?

Agentic AI Engineering

No dates yet

We will let you know as soon as the next date is scheduled.