Skip to content
Next cohort · Limited seats

Agentic Engineering

Build faster with AI agents without lowering your engineering standards.
For engineers who already vibe-coded and want the discipline back: plan, delegate to agents, and keep tests, reviews, and quality gates in place.

CourseAgentic Engineering
Designed for engineers shipping atstartups, scale-ups & AI-native teams
Starts September 16, 2026
6 Weeks

€1,500

Leave your details and we'll reach out to schedule a no-commitment call.

We only use your details to contact you about this course. Privacy policy

CourseAgentic Engineering

Course Overview

Direct AI agents to ship trustworthy software in 6 weeks

Start Date

September 16, 2026

Duration

6 weeks

Format

Online · Post-work · 2 sessions per week

Investment

€1,500

Language
English
Instructors
2 practitioners
Payment
Single payment
Certificate
Yes, upon completion

What you will be able to do

By the end of the course, you'll have done all of this, for real:

Direct agents on real projects

Use coding agents effectively on non-trivial software, not just toy examples.

Write specs agents can execute

Break large projects into agent-compatible tasks with clear acceptance criteria.

Build verification loops

Tests, evals, linters, type checks, CI, and red-team workflows that reject bad output.

Spot jagged failures

Identify where agents will silently misunderstand intent and harden the design before they do.

Review AI-generated code

Catch security holes, bad data models, and overengineering before they hit production.

Ship a defended project

A working product with a spec, evals, security review, red-team report, and engineering defense.

Is this course for you?

Self-qualify before you enroll, takes under a minute.

Prerequisites

  • Senior or mid-level software engineer
  • Confident with Git, tests, and CI
  • You've used Cursor / Claude Code / Copilot in the last 6 months (helpful, not required)
  • You bring a non-trivial repo or are willing to start one

Not for you if

  • You want to ship an autonomous agent product, pick AI Agents in Production
  • You're new to programming, pick the Advanced AI Solutions
CourseAgentic Engineering

Curriculum. Six weeks. One real project.

From vibe coding to engineering discipline with agents in the loop.

Week 01

From Vibe Coding to Agentic Engineering

Software 1.0, 2.0, 3.0. Agents as stochastic interns. Build the same idea twice, vibe-coded vs spec-driven, and make the quality gap visible. Deliverable: side-by-side write-up of vibe vs spec build with measured quality delta.

Claude CodeCursorSpec-driven dev
Week 02

Specs, Context, and Control

Prompting isn't enough. Engineer the work environment: agent-readable repo guides, executable docs, and task templates that survive contact with reality. Author Claude Code skills (SKILL.md), .claude/settings.json hooks, and at least one MCP server scoped to your repo. Deliverable: AGENTS.md, three task templates, and one repo-scoped MCP server for your final project.

AGENTS.mdCLAUDE.mdSkillsHooksMCP serversRepo guides
Week 03

Verifiability, Evals & the Quality Bar

Tests as reward signals. Build evals, golden datasets, type checks, and CI guardrails so the agent's output can be rejected before it reaches main. Deliverable: golden dataset, LLM-as-judge eval, and a CI workflow that blocks regressing PRs.

Vitest / pytestCI / LintersLLM-as-judge
Week 04

Jagged Intelligence & Human Judgment

Where agents fail and why: hidden assumptions, leaky abstractions, security mistakes. Review AI-generated code and ship a risk register. Deliverable: AI-generated code review report and risk register on your repo.

AI code reviewRisk registerThreat modeling
Week 05

Multi-Agent Workflows & Agent-Native Infrastructure

Planner, builder, tester, reviewer, red-teamer. Coordinate roles, sandbox tools, and design infrastructure agents can operate without breaking things. Deliverable: planner/builder/reviewer pipeline running on your repo with sandboxed tools.

Planner / Builder / ReviewerSandboxingTool permissionsMCP
Week 06

Final Build, Red Team & Engineering Defense

Ship a project agents helped build. Defend it with tests, evals, security review, and a red-team report. Demo Day with peer and instructor critique. Deliverable: defended project with self red-team checklist and one peer red-team review.

Demo DayRed teamPortfolio
Expert Instructors

Meet Your Instructors

Learn from industry leaders and PhD holders with extensive real-world experience

João Reis

João Reis

Co-Founder & CTO @ Medtiles, PhD AI @ FEUP

Luis Dias

Luis Dias

AI Solutions Architect @ TUI, PhD AI @ FEUP

Straightforward pricing

No "contact us" pages. Tiers auto-promote as dates pass.

Early BirdClosed
€990

Until 31 Aug

15 seats · 15 seats only

TransitionClosed
€1,250

1 – 10 Sep

Limited seats

StandardCurrent price
€1,500

From 11 Sep

Standard enrollment

TUTAI alumni
€1,100
Email us, we'll sort you out.
Company pack
€1,200/person
3+ seats from the same team.
Referral
€100credit
Bring a friend, get credit.

DGERT-certified training entity: individuals can deduct 30% of the fee on their IRS, and for companies the hours count toward the mandatory 40 hours of annual employee training.

Join Our Community

Join Our AI & Machine Learning Community

Connect with Generative AI enthusiasts, get exclusive updates, and unlock special benefits by joining our vibrant Slack community

Exclusive Events

Access workshops, webinars, and networking sessions

Connect & Collaborate

Meet AI professionals and build your network

Early Bird Discounts

Get first access to course discounts and special offers

Ask & Learn

Get help from instructors and fellow learners

Join 100+ AI Professionals

Get instant access to our community, upcoming events, and exclusive course discounts. It's free and takes just a second!

Join Slack Community

Early members get exclusive access to beta features

MK
AS
JL
RF
TC
Join 100+ members already learning together

Questions worth answering

Common questions about Agentic Engineering