Frequently asked questions

Harness Institute trains engineering, product and design teams to use AI coding agents on their existing codebases.

One method, Delegate, Review, Own, is delivered across the leading tools:

• Claude Code: claudeworkshop.com
• Cursor: cursorworkshop.com
• Codex: codexworkshop.com
• Devin: devinworkshop.com
• Gemini: gemini-workshop.com
• GitHub Copilot: githubworkshop.com

• Workshop: hands-on training delivered on your own repository.
• Pulse: ongoing monthly advisory after the workshop.
• Live cohort for individual developers, in two tiers: Simple and Applied.

The price of a workshop depends on the number of participants, the number of days and the format.

• Standard format: two days, ten to thirty participants.
• Minimum engagement: eight participants.
• Fewer than eight people? Join a cohort.
• The first month of Pulse is included.
• Travel and venue are quoted separately.

After a 30-minute call you receive a written proposal with the scope, the schedule and the price. You can also send a message with your team size and budget.

Pulse is a monthly advisory service that keeps your team current as Claude Code, Cursor and Codex change.

• A shared Slack or Teams channel with both founders, with a first response within two business days.
• A written briefing every two weeks, with a recommendation per item: adopt, test or ignore.
• A two-hour live session each month for up to ten engineers.

A three-month minimum, then cancellable monthly. If the first session does not produce three changes your team will adopt, the month is refunded.

The approach that works for the teams we train:

• Live sessions via Zoom on the team’s own repository.
• One shared rules file in the repository, so every engineer’s agent follows the same standard.
• Short exercises on real tickets, reviewed as a group.
• Regular follow-up, as the tools change every month.

On evidence, rather than by re-reading every line:

• A specification is agreed before the agent writes code.
• Each change is small enough to review in a single session.
• CI and an independent agent session check tests and standards.
• A named engineer reviews the evidence and owns the merge.

We measure against metrics your team already records:

• Time from pull request to merge.
• Defects reaching production.
• Cycle time per ticket.
• The share of AI-assisted changes shipped with tests.

We do not commit to a fixed productivity percentage.

Yes. Product and design participants learn to specify work an agent can execute, build prototypes and assess the output. Mixed groups with engineers are recommended.

• Delivery takes place under NDA, within your environment and with tools you have already approved.
• Data remains within your systems.
• Actions that require human approval are agreed in advance.

• Duration: two days as standard, one day for a narrower scope.
• Group size: ten to thirty participants, with a minimum of eight.
• Fewer than eight people? Join a cohort.
• Delivery: onsite, offsite or remote via Zoom.

Every workshop is delivered by the two founders of Harness Institute:

• Rogier Muller, CTO at BlueMonks Group, who runs coding agents in production daily.
• Vasilis Tsolis, partner at Cognitiv+, who advises regulated organisations on AI adoption.

Together they hold official ambassador roles for OpenAI, Anthropic, SpaceXAI (formerly Cursor), and n8n.

The team proficiency matrix is a short, free assessment that shows where your team stands. We use the results to tailor the workshop.

Schedule a 30-minute introductory call. We discuss your team, codebase and objectives, and follow up with proposed dates and a written proposal.