Agentic Everything Hackathon
Two days in which your team finds out where AI agents make sense in their own work, and builds the first ones.
- Format
- Kickoff call, then 2 days
- Group size
- Up to 12 people
- Investment
- €7,500 net, plus travel
- Languages
- German or English
Built for Every Team
This hackathon is for teams outside engineering: marketing, product, design, customer service, finance and leadership. You don't need to write code.
An AI agent doesn't just answer questions in a chat window. You give it a task, and it works through the steps on its own – reading, looking things up, using your tools – and hands you the result to check.
The Problem
Everyone in your company knows they should do something with AI. Almost nobody knows where to start.
So the licences get bought. People use a chat window to polish emails and summarise meetings. And the real work, the workflows that eat hours every single week, stays exactly as it was.
That is not unwillingness. Nobody has shown your people how to look at their own work and see where an agent fits and where it doesn't.
Without that you get one of two results. Nothing happens. Or someone hands you a list of fifty use cases that nobody ever builds.
The Shift
You don't start with the tool. You start with the workflow.
Map it step by step. Which steps are hard? Which are tedious? What information is needed at each step, and how long does it take?
Once that is on the table it becomes obvious where an agent helps, where a simple automation is enough, and where a human has to stay.
The people who do the work are the ones who can see this. They only need a way to look, and the experience of having built one themselves.
What Teams Build
A few examples of what comes out of two days. Yours will look different, because they start from your own workflows.
- Marketing: Turning a transcript or a YouTube video into several formats automatically – blog post, newsletter, social posts – in your own voice. A weekly check of what changed on competitors' websites.
- Product and UX design: Sorting customer feedback into themes and turning it into a first proposal for what to work on next. The same works for usability test notes and user interviews.
- Customer service: Finding the right knowledge base article for an incoming request, so the answer starts from the right place. Sorting incoming emails and routing them to the right team.
- Finance: Preparing the month-end close: the variance comments from the numbers, and a list of missing receipts.
- Leadership: A weekly briefing from team updates, meeting notes and key figures. Preparing one-on-ones from the last notes and open topics.
Before We Start
The hackathon begins with a kickoff call, about two weeks before.
Everyone gets a Miro template and maps their own workflows on it: the steps, the hard ones, the tedious ones, the information each step needs and the time it takes. Plan about two hours per person for this.
Everyone brings two real problems. We use day one to solve one of them together and day two for your own.
The Two Days
Day 1 - One Problem, Together
We take one mapped workflow from the room and solve it together.
-
Where everyone stands
A quick calibration of how your team works with AI today
-
What agents are, and what they aren't
Where the line to a chat window runs
-
Taking one workflow apart
Step by step: where an agent, where an automation, where a human
-
Live Build
We build the solution in the room, from the map to something that runs
By the end of day one, everyone has seen the whole way from a mapped workflow to a working agent.
Day 2 - Your Problem
No demo. Your work.
-
Pick your problem
One of the two you brought, checked against what we learned on day one
-
Build your prototype
Alone or in pairs, with me coaching across the room
-
Presenting results and lessons learned
What worked, what broke, what you build next
You leave with a working prototype for a problem from your own desk.
What You Leave With
- Your real workflows on a map, with the steps marked where agents make sense
- Working prototypes, built by your people, for their own problems
- A method your team can repeat on the next workflow without me
- An honest idea of where agents help and where they don't
That last point is deliberate. I run a tech organization and I have to live with what gets rolled out. I won't tell you that you need a fleet of agents and thirteen skills. Most teams need a few that fit.
Why a Hackathon
Most AI trainings end with slides and a good feeling. Then Monday comes and nothing has changed.
A hackathon ends with something that runs. People believe what they have built themselves, and they know how to build the next one.
Who This Is For
Teams of up to 12 people, from one discipline or several.
Leaders belong in the room. If you want your team to work with agents, you need to have built one yourself.
If you are an engineering team that wants a shared way of building software with agents, the Agentic Engineering Workshop is the right format.
You should bring: a laptop, access to an AI tool, and your two mapped problems. I'll send a setup checklist a week before so we spend the first morning building.
This is not an intro to prompting. It's for teams that are past "should we use this" and stuck at "where do we use this, and how".
About Daniel
I'm Daniel Hauck. At Europe's largest music retailer I'm responsible for the company's entire AI transformation, from first pilot to organization-wide rollout. On the side I build and ship products solo, using agentic workflows daily. I've run the Agentic Engineering Workshop over 10 times with nearly 100 professionals across tech, product and marketing. The hackathon brings that system to teams outside engineering, in a form I've already run with several different teams.
Logistics & Investment
Format: Kickoff call, then 2 days on-site at your office or remote
Group size: Up to 12 people, to keep it hands-on. Smaller groups welcome, reach out and we'll find a setup that fits.
Investment: €7,500 (net; VAT added where applicable), plus travel. Kickoff call, Miro template and setup checklist included.
Languages: German or English
One offer, no tiers. If the format doesn't fit your team, we talk and adjust it before you commit.
What Happens After
The hackathon gets your team through it once, together. The goal is that they keep going without me.
From there the work is making it normal: the next workflows, shared standards, and a way to get everyone to the same level.
If you want help with that inside your company, we can talk. It usually works best by starting with one team and going from there.
What participants say
Participants from outside engineering on two days of building with AI agents.
“What I valued most: we worked on real projects, not on artificial exercises.”
Emmy
Organizational Developer
“I really liked the format: on day one we built a project together, on day two everyone worked on their own. That got us up to speed quickly and gave us enough knowledge to actually build something ourselves the next day.”
Paul
Projekt Manager
Questions
Which AI tools do we need, and who provides the licences?
Ideally the ones your company already uses. In the free call we look at what you have. If nothing fits yet, I'll recommend a tool, for example Notion or Langdock. Your company provides the licences, and everyone needs working access before day one. The setup checklist I send a week before says exactly what's needed.
How much preparation does it take, and do we need a Miro account?
Plan about two hours per person, spread over the two weeks between the kickoff call and the hackathon. That time goes into mapping your own workflows on the Miro template and choosing your two problems. You work in your own Miro: I send you the template, and you copy it into your workspace. If you don't use Miro yet, a free account is enough or use Confluence, Notion etc.
What data can we work with? Is anything sent to AI providers?
Whatever your company's rules allow in the tool you use. You work with your own licences, under the terms your company has agreed with the provider. That includes Miro: you map your workflows in your own workspace. I don't add any tools, and no data goes to any other third party. For personal data, such as customer or employee records, test or anonymised data is usually the better choice for a first prototype. We clarify this in the kickoff call.
On business and enterprise plans, the major providers don't use your data for training by default. It's still worth checking your provider's privacy settings beforehand.
What happens to the prototypes afterwards?
They're yours. They're built in your tools, with your licences, and stay in your company. Your people know how they work because they built them, so they can keep developing them. Which ones you roll out more widely is your decision.
Our team has very mixed experience with AI. Does that work?
That is the normal case. Day one starts with a calibration of where everyone stands and then takes the whole group through one workflow together. On day two everyone works on their own problem, at their own level.
We are more than 12 people. Can we still do this?
Yes, in several groups. The cap of 12 is what keeps the two days hands-on, so a larger team runs the hackathon more than once. It usually works best to start with one team and go from there.
How far ahead do we need to book?
A few weeks' lead time helps, also so you can prepare yourselves and your team. For the hackathon, plan at least four weeks, since the kickoff call takes place about two weeks before.
Do participants get a certificate?
Yes. Everyone gets a certificate of attendance with the content and duration of the workshop.
What if we need to cancel or reschedule?
One rescheduling is free if you let me know more than two weeks before the original date, subject to availability. For cancellations: more than four weeks before the date, it's free; two to four weeks before, 50% of the agreed fee; less than two weeks before, 100%. The binding terms are in your offer.