Avanscoperta
The Agentic Developer Masterclass

The Agentic Developer Masterclass

Stop prompting. Start engineering with intent.

Nobody knows where AI development will be by the time you read this. The tool you master today may be deprecated next quarter, which is exactly why this Masterclass is built around harness engineering: the principles you’ll keep using long after the tools change.
The destination is harness engineering: building and tuning the whole system around the model so it produces reliable work, stays inside your token budget, and survives the next model release.

In this two-day workshop with Marco Heimeshoff, we’ll cover the practical end (CLAUDE.md, skills vs agents, forked vs clean subagents, fighting context rot) and the serious end (agent teams, observability, EU AI Act compliance, local LLMs on an open harness for privacy-sensitive work). And underneath all of it sits the why: modelling and intent, where 15 years of DDD practice turns an agent from a fast typist into a pair that understands what the software is for.
You will leave with a working harness for your own projects and the principles to keep adapting it as the ground keeps shifting.

This Workshop is co-organized with KanDDDinsky Conference. 💚

Learning Objectives

By the end of this Masterclass you will be able to:

  • Trace the progression from prompt engineering to context engineering to harness engineering, and recognise which problem each layer actually solves so you stop reaching for the wrong one.
  • Set up the anatomy of Claude Code (and Codex): CLAUDE.md, skills, agents, commands, plugins and hooks composed into a harness that holds up across long sessions and real codebases.
  • Fight context rot and keep token usage slim: forked vs clean subagents, when to pick an agent over a skill, and the trade-offs of each.
  • Apply the core agentic working patterns (prompt chaining, routing, parallelization, orchestration, evaluator-optimizer, the bookend pattern) and know when to reach for agentic vs programmatic workflows.
  • Decide between MCP and CLI usage, wire up hooks correctly, and run agent teams without losing control of cost or context.
  • Compare the existing agentic frameworks (Agentheim, BMAD, GSD, NWAVE) against your own harness rather than adopting one wholesale.
  • Build memory and observability that fit your situation, from MD files and Obsidian up to enterprise systems, including compliance constraints (EU AI Act) and privacy-sensitive setups on local LLMs with an open harness.
  • Use the domain-driven loop (modelling and intent) to give agents the why, so the software they produce matches what the business actually meant.

The Trainer

Marco Heimeshoff is a Software developer, Domain-Driven Design trainer, coach, and consultant who has worked with organizations ranging from Intuit, Lego, Mercedes, Spotify, and Starbucks to hundreds of smaller companies.

Marco co-founded the KanDDDinsky conference and VirtualDDD.com.
His development philosophy centers on lifelong learning, focus on language, and empathy.

Target Audience

This is an advanced masterclass, but you don’t need to arrive as an expert. It is aimed at developers, tech leads and architects who want to build agentic systems they can depend on rather than one-off demos. If you have solid development experience, two days are enough for us to bring you up to speed, whatever your starting point with agents.

You will get the most out of it if you:

  • Code regularly and are comfortable in a terminal (no shell wizardry required, but `cd`, git and editing config files should not slow you down).
  • Bring a laptop with Claude Code and/or Codex installed and an active subscription before day one. It is in person, so there is no time to debug installs on the day.
  • Prior experience with Claude Code, Codex or a comparable agentic CLI is a bonus and will let you go further faster, but it is not a requirement. If you have already hit the wall where naive prompting stops scaling, you’ll recognise exactly what we’re solving.

Topics

  • From prompt engineering to context engineering to harness engineering: the through-line.
  • Anatomy of an effective Claude/Codex setup: CLAUDE.md, skills, agents, commands, plugins.
  • The three loops: building software (what), the harness (how), domain and intent (why).
  • Agentic working patterns: chaining, routing, parallelization, orchestration, evaluator-optimizer, bookend.
  • Context discipline: preventing context rot, token saving strategies, subagents, agents-vs-skills, MCP-vs-CLI, hooks.
  • Workflows and frameworks: agentic vs programmatic, and reading Agentheim / BMAD / GSD / NWAVE against your own harness.
  • Governance and reach: memory systems, transient software, observability, EU AI Act, local LLMs on an open harness.

Agenda

Two full days, in person, run as one continuous build: roughly 9:00 to 17:00 each day with lunch and two coffee breaks. The room works a shared wall of physical stickies, and the harness each participant starts on day 1 is the same artifact they keep extending through day 2.

Day 1 – First Contact, and Why the Harness is the Point

The coding loop and the progression

  • Run Claude Code, feel the coding loop, read `/context` and use `/clear`.
  • The progression as a story: prompt → context → harness, and the three loops as the map for the whole workshop.
  • Reality check: the METR slowdown finding vs structured-workflow gains (same model, different harness, opposite outcome).
  • Your first real CLAUDE.md: the seed of the harness you extend all workshop.

Build the harness: anatomy of an agentic setup

  • Prompt vs context vs harness engineering, with the historical progression and why each layer appeared.
  • Anatomy of the setup: CLAUDE.md, skills, agents, commands, plugins and how they compose.
  • The trade-offs: context rot prevention, token saving strategies, forked vs clean subagents, agents vs skills, MCP vs CLI, hooks for deterministic enforcement.
  • Memory systems as a layer, from MD files and Obsidian up to enterprise.

Day 2 – The Harness in the Wild, and the Domain-Driven Why

Patterns, workflows, frameworks

  • The named agentic working patterns: prompt chaining, routing, parallelization, orchestration, evaluator-optimizer, the bookend pattern, and when not to over-engineer.
  • Workflows: agentic vs programmatic, agent teams, and hooks as safety rails inside the workflow.
  • Framework bake-off: Agentheim, BMAD, GSD, NWAVE (what each opts into, where they agree, how to read one against your own harness).

Domain-driven harnesses: intent, compliance, transient software

  • The domain-driven loop: modelling and intent as the why behind the coding and harness loops; ubiquitous language to reduce the ambiguity LLMs amplify.
  • Interview-driven spec work as collaborative discovery, the Event Storming parallel this room already knows.
  • Transient software: when a build is cheap enough, throwaway solutions become a legitimate design choice.
  • Observability and EU AI Act compliance, plus privacy via local LLMs on an open harness. Swap the model, keep the harness.

Why you should join the Masterclass

Most courses teach you to use a tool. In the Agentic Developer Masterclass, you will learn to design the system around it, a different and more durable skill. Over two intensive, hands-on days we follow one spine (prompt → context → harness) through three loops: building software (the what), the harness itself (the how), and domain and intent (the why).
Along the way you will learn every component a dependable harness is made of, from memory and context management through workflows, tools and skills, and how they compose into a system you can actually rely on.

About the workshop

Venue: park inn, Alexanderplatz 7, 10178 Berlin Germany
Language
: English
Time: from 9.30 am to 5.30 pm each day
Max number of participants: 16 people
Dress code: Informal. We recommend you wear comfortable shoes as the majority of the activities will involve walking around and standing up most of the time
Included in the price: arrival coffee break, lunch and afternoon coffee break, each day
Not included in the price: travel and accommodation.

FAQ

Is this a beginner course on AI coding?
It’s an advanced masterclass, but you don’t need to arrive as an expert. If you have solid development experience, two days are enough for us to bring you up to speed, whatever your starting point with agents. Prior time in an agentic CLI will let you go further faster, but it isn’t a requirement.

Do I need my laptop?
Yes, and you need it ready before day one. Bring a laptop with Claude Code and/or Codex installed and an active subscription. It is in person, so there is no time to debug installs on the day.

Do I need to know DDD?
Not in general. But this run is the KanDDDinsky pre-conference workshop, so the domain-driven loop assumes a room fluent in EventStorming, bounded contexts and ubiquitous language, and goes deeper because of it.

Will the tool details still be current by autumn 2026?
Some won’t, and that is the point. We lean on durable principles over tool trivia by design, so you keep adapting your harness as the ground shifts rather than relearning from scratch.

Registration logos kanDDDinsky and avanscoperta

This workshop is co-organised with KanDDDinsky, and all registrations will be handled by the KanDDDinsky team via their Pretix platform.

Book it for your company
Contact us using the form at the bottom of this page if you wish to arrange a private custom edition of the EventStorming Master Class.

Vuoi saperne di più sui nostri biglietti?

Biglietti

UNDERCOVER_LEARNERSCONTO 20%

Se ti stai iscrivendo al corso a titolo personale e non hai la Partita IVA, c’è una tariffa speciale per te: uno sconto del 20% sul prezzo corrente del biglietto, che potrai ottenere inserendo il codice UNDERCOVER_LEARNER prima di procedere al checkout.

N.B.: Il codice non può essere utilizzato con nessun’altra offerta, ed è valido su tutti i tipi di biglietto.

Marco Heimeshoff

Marco Heimeshoff è Software Developer, Trainer di Domain-Driven Design, facilitatore di EventStorming e Speaker.

Il suo focus attuale è lo sviluppo agentivo con AI: aiuta sviluppatori e team ad andare oltre il vibe coding e a costruire workflow AI-augmented consapevoli e sostenibili. In questo campo porta una prospettiva unica — oltre 15 anni di esperienza pratica nella gestione della complessità attraverso il DDD, ora applicati alla progettazione di ambienti di sviluppo agentivi. È anche l'autore di agenthoff, un opinionated Domain-Driven Agentic harness, e creatore di WhisperHeim.

About us

Avanscoperta is a growing community of professionals learning together, exchanging experiences and ideas, exploring uncharted territories.

We carefully select the most relevant topics and the best trainers and thought leaders worldwide with an eye to the most engaging teaching techniques.
Are you ready to learn with us?

Ti interessa questo corso?

Se hai delle domande specifiche, o vuoi ricevere maggiori informazioni, scrivici. Saremo felici di risponderti!







    NEWSLETTER

    Vuoi ricevere una volta a settimana i nostri contenuti esclusivi su tecnologia, sviluppo software, business e design?

    ISCRIVITI ALLA NEWSLETTER :-)

    Iscriviti alla newsletter!