Private team workshop · on your own stack

In one day, your engineers ship working AI agents on your own network.

Not a lecture, not slideware. Up to 12 of your people leave with running agents built on your gear — or you don't pay.

Book a 15-minute fit call → 📄 Send it to your manager

60 practitioners at our sold-out AutoCon5 Munich workshop · Heard on Packet Pushers
On-prem · air-gap capable · your data never leaves the building

Honest fit

Built for regulated NOCs and SOCs that can't send data to the cloud.

This is for you if…

You run an enterprise NOC/SOC with dedicated network or security engineers · you already have at least one automation initiative in production (Ansible, Python, ITSM) · your board is asking for an "AI plan" and you need something real · your data has to stay on-prem.

It's not for you if…

You want generic "AI literacy" or a ChatGPT-for-business session · you expect autonomous production changes with no human-in-the-loop · you have no engineers to own the system after the day. We'll tell you honestly on the call.

What your team builds

Three working agents — on your infrastructure, not a demo sandbox.

By the end of the day these are running against your own devices and workflows. Your engineers built them, so your engineers can keep building them.

Alert-triage agent

Cuts the night-shift noise down to the handful of alerts that actually need a human — using your alert sources and your severity logic.

Config-generation agent

Turns intent into vendor-correct configuration your team reviews and ships. Multi-vendor, grounded in your standards.

Alert-to-action agent

Closes the loop from signal to a validated, approved change — with the human-in-the-loop gate you decide.

The day

A working agenda, not a webinar.

Day 1 €4,900 · included in both tiers

  1. Ground rules & the on-prem AI stack — what runs where, and why nothing leaves the building
  2. Connect to your environment (lab or read-only prod) and stand up the models locally
  3. Build the alert-triage agent on your real alert data
  4. Build the config-generation agent against your vendors and standards
  5. Wire the alert-to-action loop with a human-in-the-loop gate
  6. Hand-off: your team demos the agents they built

Day 2 €7,500 · the deeper build

  1. Harden the agents against your edge cases and failure modes
  2. Add digital-twin pre-validation so changes are tested before approval
  3. Fit the deterministic execution catalogue to your change process
  4. Tiered governance: what runs automatically, what waits for sign-off
  5. Map the path from workshop to a scoped 90-day AutoNOC pilot
  6. Leave with a written next-step plan your board can read

What you leave with

Running software and a team that can extend it.

✓ 3 working agents on your stack ✓ The code, in your hands ✓ Your engineers trained to extend them ✓ On-prem deployment pattern ✓ A written path to a pilot ✓ No vendor lock-in

Risk reversal

If your engineers don't leave with working agents, you don't pay.

The workshop is built on your own stack, so "it works" means it works where it counts. We carry the risk — you don't write a cheque for a maybe.

Real constraint

Two workshops a month. That's it.

Each one is built for your environment before we arrive — that prep is why we cap at two. Next open slot: limited.

The bridge

Your fee credits 100% toward a pilot.

Sign a 90-day AutoNOC pilot within 60 days and the full €4,900–€7,500 comes off the price. The proof pays for itself.

Before you ask

Security, air-gap, prerequisites, team size.

What's the security posture? Does anything leave our network?+
No. The models are open-weight and run on your hardware. The workshop is designed for on-prem and air-gapped environments — your configs, telemetry and topology never leave the building. Nothing is sent to a cloud API and nothing becomes training data for anyone else.
Can we run it fully air-gapped?+
Yes. We bring the model weights and tooling in; we work against a lab or a read-only view of production, whichever your policy allows. If you're fully air-gapped, we plan the offline setup with you before the day.
What are the prerequisites?+
Engineers comfortable with the CLI and some Python, a machine or VM we can run local models on (GPU helps, not always required), and access to a lab or representative device data. We send a short prep checklist once a slot is booked.
How many people, and what roles?+
Up to 12. Best mix is network/SOC engineers plus one automation lead. It's hands-on — everyone builds, nobody just watches.
1 day or 2 — which do we need?+
One day proves the three agents on your stack. Two days hardens them, adds digital-twin pre-validation and governance, and maps the pilot. If you're heading toward a pilot, take two.
What does the guarantee actually mean?+
If your engineers don't walk out with working agents running on your infrastructure, you don't pay. It's that simple — the risk is ours.

For the person who signs off

Send your manager the one-pager.

Price, scope, guarantee, security posture and outcomes on a single page — everything they need to say yes.

📄 Get the one-pager on a 15-min call →

One call. Fifteen minutes.

Tell us your environment. We'll tell you if a workshop fits.

No deck, no interrogation. You describe your stack and what your board is asking for — we tell you whether a workshop, a pilot, or honestly someone else is the right call.

Book a 15-minute fit call →

15 minutes · no obligation.