NetPilot

AI for Enterprise Networks

A custom AI agent, tailored to your use case and wired into your workflow — from network design to change validation — built for you with a 100% success guarantee.

app.netpilot.io

Built on real multi-vendor network OSes — not approximations

9+
network OSes and growing, on real CLIs
~2 min
from prompt to a running multi-vendor lab
Unlimited
AI + lab usage for your agreed scope
4 ways
to deploy — cloud, dedicated, on-prem, air-gapped

The category, defined

What is enterprise network simulation software?

Enterprise network simulation software lets a team build and test a faithful multi-vendor network off production — to validate a change, run a POC, or research a protocol. NetPilot is the AI-native version: describe the network in plain English and, instead of hand-building every node like Cisco CML, EVE-NG, or GNS3, an AI agent designs the topology, writes the per-vendor configs, and deploys real network-OS containers (true emulation, not approximations) across every device in parallel — about two minutes from prompt to a running lab.

The agent hands back a consolidated change report, and direct CLI access over SSH is always there to verify its work. It is a Day-0/Day-1 runnable mirror — it never touches production — delivered founder-led (Signature) or dedicated on-prem (Enterprise).

AI-native — the agent builds itReal NOS, not approximationsDay-0/Day-1 · never touches productionCloud · dedicated · on-prem · air-gapped

Deploy on your terms

The same platform, wherever it has to run — from a two-minute cloud lab to a fully air-gapped install in your data center.

Shared cloud

Self-serve in the browser — a dedicated VM per lab and a running multi-vendor topology in ~2 minutes. Where Signature delivery runs.

Dedicated environment

Isolated, dedicated resources for your team — no shared capacity, scoped to your workloads.

Self-hosted / on-prem

Run NetPilot in your own data center or private cloud, behind your perimeter.

Air-gapped

Fully offline via a private registry for regulated or disconnected environments.

Your choice of AI provider per deployment — Claude, OpenAI, Gemini, or AWS Bedrock — for your data-residency and procurement requirements.

Built for the whole org, not a metered seat

Unlimited usage for your scope, wired into the tools you already run, with any network OS your team needs.

Unlimited usage

No credit caps, no per-lab metering — run every multi-vendor lab your agreed scope needs, where Cisco CML, EVE-NG, and GNS3 cloud meter you by node count or credits.

Wired into your stack

MCP / API into NetBox, ServiceNow, Git, and CI/CD — build a runnable digital-twin sandbox from your real network context (roadmap + hands-on delivery). A REST API spins up a fresh twin per pull request.

Any vendor NOS, built for you

9+ network OSes and growing on real CLIs; SONiC, Cisco IOS-XE, Juniper vMX, NVIDIA Cumulus, and Aruba CX are custom-built for you on the enterprise plan.

Security & data handling

The controls a security reviewer scans for — and the deployment options that keep your data where you need it.

SSO

Single sign-on for your team.

Audit logs

Who deployed what, and who ran which CLI command.

Isolated environments

A dedicated, isolated environment per lab — no shared tenancy.

Data residency

Keep data in your environment with on-prem deployment and your choice of AI provider.

BYOI — no image distribution

We never distribute commercial vendor images; bring your own licensed NOS or sanitized configs.

Never touches production

Labs are Day-0/Day-1 — a runnable mirror, isolated from your live network.

Built for your security requirements.Dedicated environments, SSO, audit logging, and on-prem deployment — tell us your specific compliance and data-handling needs on the scoping call, and we'll scope the deployment to meet them.

Two ways to work with our team

Both start with a 30-minute scoping call with the founder — hands-on delivery on our cloud, or NetPilot dedicated in your environment.

Signature

Founder-led, hands-on delivery on our shared cloud — we scope your network and build your first multi-vendor labs with you.

$500/ month

By application · limited spots · 100% success guarantee.

  • Unlimited AI & lab usage, for your agreed scope
  • Custom vendor builds + MCP/API into your stack
  • The founder builds your first labs with you
  • A money-back success milestone

Enterprise

NetPilot in your environment — dedicated and on-prem, with your choice of AI provider and the controls your security team needs.

Custom

Tailored to your network, deployment, and AI provider.

  • Everything in Signature
  • Self-hosted / on-prem (your cloud or data center)
  • Your choice of AI provider (Claude, OpenAI, Gemini, Bedrock)
  • Dedicated resources, SSO & audit logs
  • Share lab artifacts; embed in any workflow

See the full plan comparison on the pricing page.

Use cases

Built for every team

One AI-native network lab, tailored to how each team works. Pick yours.

Network

Mirror production and validate BGP and ACL changes on real CLIs before they ship.

Explore Network change validation

Mirror, snapshot, validate

Pre/post snapshot diffs capture routing tables, neighbor state, and ACL behavior before deployment.

Multi-vendor in one sandbox

Cisco, Juniper, Arista, Nokia, Palo Alto, and Fortinet in one lab; the agent writes the per-vendor syntax.

~2 minutes, not days

About two minutes from a plain-English prompt to a working multi-vendor mirror on real NOS.

Rehearse the failure

Shut a backbone link or kill a BGP peer and verify reconvergence before touching production.

Enterprise network simulation — FAQ

The questions enterprise network teams actually ask

NetPilot is an AI-native network emulation platform built for enterprise teams: an AI agent designs, builds, and validates real multi-vendor networks from a plain-English prompt, in the cloud or on-prem. Unlike traditional tools (GNS3, EVE-NG, Cisco CML) where you hand-build every node, NetPilot deploys real network-OS containers in about two minutes and writes the per-vendor configs for you. It runs 9+ network operating systems and growing, with bring-your-own-image for licensed vendor NOSes and custom builds on the enterprise plan. Founder-led Signature delivery and on-prem Enterprise deployments are available.
No. The Signature ($500/mo) and Enterprise plans include unlimited AI and lab usage for your agreed scope — no monthly AI-credit cap and no per-lab metering. This is the key difference from credit-capped tiers on platforms like Cisco CML, EVE-NG, or GNS3 cloud: you can run as many multi-vendor labs, as often, and as large as your scope requires. Enterprise on-prem runs on your own infrastructure, so capacity is bounded only by your hardware.
NetPilot is MCP-connectable by design. The roadmap wires the agent into your source of truth (NetBox / Nautobot inventory, Git config repos) and ticketing like ServiceNow so a runnable digital twin is built from your real topology, not hand-typed. Today, on Signature and Enterprise, the team builds a faithful twin by ingesting sanitized configs or describing the segment in plain English, and the agent generates the matching multi-vendor runnable mirror in minutes. The DIY alternative is generating ContainerLab YAML from NetBox by hand; NetPilot's agent writes and deploys it for you. MCP / API integration into NetBox, ServiceNow, and your stack is part of hands-on Signature delivery and Enterprise workflow embedding.
Yes. NetPilot is built on ContainerLab, so any image that runs in ContainerLab can be deployed. Nokia SR Linux, FRR, and Linux are built in on every plan; Cisco IOL, Juniper cRPD, Arista cEOS, Palo Alto PAN-OS, and Fortinet FortiGate are available via bring-your-own-image (BYOI). SONiC and other custom or exotic NOS images — Cisco IOS-XE, Juniper vMX, NVIDIA Cumulus, Aruba CX — are built for you on the enterprise plan with a dedicated environment. The supported lineup is 9+ and growing.
Yes, via NetPilot's Enterprise tier — deploy in your own data center, private cloud, or air-gapped via a private registry, with your choice of AI provider (Claude, OpenAI, Gemini, or AWS Bedrock) per deployment. For fully offline regulated environments where you must own every layer, DIY ContainerLab, EVE-NG, or GNS3 on your own hardware remain valid choices; NetPilot Enterprise covers cloud-managed on-prem and air-gapped installs as a managed enterprise option with dedicated environments, SSO, and audit logs.
NetPilot is built for enterprise security requirements: SSO, audit logs (who deployed what and who ran which CLI command), dedicated and isolated environments per lab, and data residency via on-prem or air-gapped deployment with your own choice of AI provider. We never distribute commercial vendor images — you bring your own licensed NOS or sanitized configs (BYOI) — and labs are Day-0/Day-1, isolated from your production network. We scope your specific compliance and data-handling requirements on the call rather than implying certifications we don't hold.
Signature ($500/mo, by application) is founder-led, hands-on delivery on NetPilot's shared cloud: the founder scopes your network in a 30-minute call and builds your first multi-vendor labs with you, with unlimited usage for your agreed scope, custom vendor builds, MCP / API integration, and a money-back success milestone. Enterprise (custom pricing) adds NetPilot in your environment — self-hosted / on-prem, dedicated resources, your choice of AI provider, shared lab artifacts across the team, and embedding into any workflow (change management, research, training, pre-sales). Signature is “NetPilot's team builds it with you”; Enterprise is “NetPilot runs in your environment, dedicated to your team.”
The scoping call is a 30-minute session with the founder to understand your network and the outcome you need. You walk through the topologies, vendors, and workflows you want to lab — change validation, POC demos, research, or training — and the founder defines the agreed scope, identifies any custom vendor images to build, and maps the MCP / API integrations into your stack. For Signature, it sets the success milestone tied to the money-back guarantee; for Enterprise, it scopes the on-prem / dedicated deployment and AI-provider choice. It is a working session, not a sales demo — you leave with a concrete build plan.
On the Signature plan, the founder agrees a concrete success milestone with you during scoping — a working multi-vendor lab, a validated change, or a delivered POC, defined in advance. The work is hands-on: the founder builds it with you, not just advises. If that agreed outcome isn't delivered, you get your money back for that milestone. It removes the risk of paying for a hands-on engagement before you've seen it produce the result you came for.
Yes. A NetPilot lab is a runnable digital twin — the staging surface for any automation that pushes network changes. Run Ansible playbooks, Python with Netmiko / Nornir / NAPALM, Terraform network providers, or NETCONF / RESTCONF against the twin's real NOS — the same IOS / JunOS / EOS / SR Linux syntax as production. A REST API lets CI/CD spin up a fresh twin per pull request and tear it down on merge, so automation is validated on real device code before it touches the live network.
A network digital twin comes in two forms: continuous / telemetry twins that read-only model the live production network (Forward Networks, Ciena, Nokia, VIAVI), and runnable twins you build on demand and actually execute changes on. NetPilot is the on-demand, runnable twin — not a continuous telemetry-fed twin. Batfish and Forward do model-based or formal verification (“will the BGP session come up?”) without executing the network; NetPilot deploys real NOS, applies the change, watches OSPF reconverge, and you SSH in to verify. They are complementary: use the formal model to prove correctness, the runnable mirror to watch actual behavior. NetPilot owns the Day-0/Day-1 build-and-validate lane and never touches production.
Yes, on the Enterprise plan you choose the AI provider per deployment — Claude, OpenAI, Gemini, or AWS Bedrock — so the agent runs under your preferred model, data-residency, and procurement terms. This matters for regulated and on-prem environments where the model endpoint and data handling must meet internal policy. The shared-cloud Signature plan runs on NetPilot's default model.

Explore the platform

Network Digital Twin

Build a runnable multi-vendor replica and execute real changes before production.

Network Change Validation

Pre/post test BGP, ACL, and routing changes on a runnable twin.

POC Network Lab

Stand up customer-shaped multi-vendor demos in minutes for sales engineering.

Network Research Lab

Reproduce cross-vendor behavior and inject failures for protocol R&D.

ContainerLab in the Cloud

ContainerLab's real-NOS model, with the AI agent authoring the YAML and configs.

Cisco CML Alternative

Multi-vendor, AI-built, and cloud-or-on-prem — beyond Cisco-only labs.

Tell us what you need to lab

A 30-minute scoping call with the founder — we'll map your network, scope the work, and (on Signature) set a money-back success milestone.

See how the plans compare on the pricing page.