Service: AI Engineering Practice
FOUR TO SIX WEEKS / PRACTICE ENGAGEMENT

AI coding adoption with Claude Code and agent workflows.

Build an AI engineering practice your whole team can use.

We help a team already using Claude Code or Cursor define its agent workflows, review process, evaluation suites, and internal playbook. The team gets a repeatable process with an audit trail, ready for wider use.

Book a fit call
Run an AI engineering audit

You receive one email containing the audit worksheet.

Starts from EUR 10k. Four to six weeks, plus an optional retainer.

01
WHO THIS IS FOR

This is for teams already shipping with coding agents.

  • 01 Your engineers are already using Claude Code, Cursor, or another agent in their daily work.
  • 02 Output has increased, but methods still differ from person to person.
  • 03 Two or three people get consistent results while the rest of the team has yet to adopt their approach.
  • 04 Your code review process does not record what an agent generated, when it generated it, or the intended outcome.
  • 05 You have heard the vendor pitches and now need a shared practice owned by your team.

When at least three apply, your team is ready to formalize its AI engineering practice.

02
WHAT YOU GET

What your team will use after the engagement.

Part 01

Agent workflows

We document the loops your team will use for code generation, refactoring, review, testing, and documentation. For each loop, we name its owner and define the input contract and exit criterion.

Part 02

Review rituals

The review rubric separates the checks an engineer performs from the rules enforced by the runtime. It also defines when to escalate an issue. Every review remains assigned to an accountable engineer.

Part 03

Evaluation suites

A lightweight test suite shows when an agent starts to regress. Its scope stays practical enough for the team to run every week.

Part 04

The internal playbook

Your engineering lead owns a written playbook that records the practice. New engineers can learn it in hours, and the document is available to the whole team.

03
HOW THE ENGAGEMENT RUNS

The core engagement runs for five weeks.

  1. Week 1

    Audit

    With your team

    We observe the team at work and separate established habits from improvisation. We then choose the three workflows worth documenting with your team.

  2. Week 2

    Codify

    With your team

    We write each workflow with the team, including its owner, input contract, and exit criterion. This turns existing habits into an explicit process.

  3. Week 3

    Review and evaluate

    With your team

    We design the review rubric and evaluation suite around the team's normal work, with a scope they can maintain.

  4. Week 4

    Embed

    With your team

    We use the new practice on active work for a week, then revise and document it with the team.

  5. Week 5

    Hand off

    With your team

    The engineering lead receives the internal playbook. The optional retainer provides monthly office hours to review the practice and adjust it as the tools or team change.

04
THE GRADUATION PATH

What happens after handoff.

You own the workflows, review rituals, evaluation suite, and playbook. They remain usable if you stop working with us or change tools.

If you continue, the team uses the practice with EHOS, which records governance and audit information for every AI-generated change. Each change arrives ready for review, while the technical owner remains accountable.

05
THE MAP
A useful practice belongs to your team and lasts beyond the tool that introduced it.

EmpoweredHouse / AI Engineering Practice

06
PRICING

Starts from EUR 10k.

The final price depends on team size, whether the work spans one repository or several, and whether compliance review sits on the critical path. The optional retainer is priced separately. Most teams use it during the first quarter.

Team size

How many engineers the practice must reach.

Surface area

One repo or many.

Compliance review

Whether compliance is on the critical path.

During the fit call, we quote the engagement and check the team's readiness. Teams without working AI workflows receive a recommendation for an earlier engagement.

07
THE FIT CALL / 30 MIN

Check whether your team is ready to formalize its AI practice.

The fit call takes thirty minutes. We examine how your team ships with AI, why it chose that approach, and which parts are repeatable. We then recommend this practice engagement or an earlier service.

Book a fit call Run an AI engineering audit