%402x.svg)
Your senior engineers still hand-write staging models, tests and docs that an agent can draft. In one day your data and analytics engineers build a dbt warehouse, Airflow orchestration, CI/CD and a Claude Code agent that turns a Jira ticket into dbt code, tests and a pull request, and keep the code and the agent setup.
The workshop at a glance
One day at your office, five build blocks from 09:00
Data and analytics engineers and their team lead, up to 12 people. Basic SQL and Git are enough, because the DAGs and the agent start from our templates
From EUR 9,000. The cloud environment for the day and travel inside the EU and the UK are included
Our cloud account, with a synthetic dataset. At the end of the day the code moves to your Git organisation
Project managers and analysts joining too? Data AI in a Day takes the whole data team through one project →
What agent-assisted delivery did at a global manufacturing group, what it can be worth for your team, and the fee
A three-layer warehouse (staging, curated, presentation) deployed from a dbt template, with naming conventions and tests wired up and passing.
A dbt project with every test passing
Airflow DAG templates and environment configs, then the first automated deploy: every change goes from branch to pull request, through a CI build with dbt build and test, to merge and deploy.
A pipeline on a schedule and CI/CD with an audit trail
An agent that reads Jira tickets, sorts them by intent and routes each type, with structured output and error handling.
The intent router and its prompt templates
An engineer writes a Jira ticket. The agent writes the dbt model, tests and documentation and opens the pull request; the team reviews and merges it, and the table is live.
Merged agent-drafted pull requests and the tables they deploy
The team works through complex tickets, schema drift and multi-table dependencies with the agent, then sizes how much of its own backlog is pattern work and writes the 30-day playbook for its stack.
A 30-day playbook with week-one tasks
Stack for the day
Claude Code, dbt, Airflow, Git-based CI/CD and Jira, or Azure DevOps on request, on our cloud account. At the end of the day the code moves to your Git organisation, we switch the environment off, and the dbt project and DAG templates connect to your own sources from there.
Which workshop fits
This day is for data and analytics engineers who want the delivery setup itself, from warehouse template to agent. Data AI in a Day is for a mixed team of project managers, data protection officers, engineers and analysts who take one project from request to merged pull request together, and the AI analytics workshop is for analysts who build agents that explain why the numbers moved.
5
EUR 180,000
40% (assumption)
60% (assumption)
About EUR 216,000
The model: 5 × EUR 180,000 × 40% × 60% = EUR 216,000. The 60% sits inside what the global manufacturing group measured, where total effort per high-complexity task fell from 11 person-days to 3.5-5. In the 15:30 block your team redoes the sum with its own figures.
Scoping call, 30 minutes
We walk your stack and your backlog, and choose the ticket types the agent takes first.
Proposal
We send the fee, the scope and the date, with the ticket types agreed in the call.
Before the day
We set up the environment on our cloud account, with a synthetic dataset.
The day, from 09:00
Your engineers work through five build blocks at your office. At the end the code moves to your Git organisation and the playbook stays with your team.
After the day
In weeks one and two we hold a follow-up call with your workshop sponsor on what landed and which tickets come next. For 30 days after the day, our engineers support yours on Microsoft Teams. In weeks three and four, if you want it, a Data Readiness Assessment, EUR 9,000 to 15,000 and credited in full towards any engagement, maps your data estate to the setup you built. After that your team runs it, or adds a monthly capacity pack.
No. The day runs on our cloud account with a synthetic dataset, so your production code and data stay outside the engagement. At the end of the day the code moves to your Git organisation and we switch the environment off.
You do, from the end of the day: the dbt project with its tests and documentation, the Airflow DAG templates, the CI/CD configuration, the agent setup with its prompt templates, and the 30-day playbook. To keep using the agent after the day, your team needs its own Claude Code licences.
Data AI in a Day trains a mixed data team, from project managers to analysts, on one project from request to merged pull request. This workshop is for engineers only and goes deeper into the delivery setup: the warehouse template, orchestration, CI/CD and the agent that works the Jira queue. Each is one day from EUR 9,000; booked together, the two days cost from EUR 18,000.
No tool does that on its own. Every change goes through branch, pull request, a CI build with dbt build and test, merge and deploy, so you can show who changed what, who approved it and which tests passed. That audit trail is part of the evidence SOX change controls ask for. We built a finance warehouse on dbt and Airflow for a global professional services group that works under SOX.
On site at your office, for up to 12 engineers from one company, so everyone gets hands-on time with the agent. Inside the EU and the UK, travel is included in the fee; elsewhere it is a separate line in the proposal.
You talk to Michał Dębski, co-founder, about your stack and the ticket types an agent would take first. If the day fits, you get a proposal with the fee, the scope and the date.
