Enterprise compliance software has a reputation for six-month implementations. KeplerAI is built so you can be scoring real data the same afternoon you sign up. Here’s the whole path.
Minute 0 — sign up
You create an account and your tenant is provisioned automatically, with RBAC roles ready (admin, business, data scientist). No ticket, no waiting on our team.
Minute 2 — install the data plane
The SaaS issues a single-use, short-lived, tenant-scoped enrollment token. You run one command in your environment:
docker run kepler-runner --enroll <token>That’s the entire install. The Runner self-registers over an outbound connection — no inbound port, no firewall exception.
Minute 5 — the Runner is online
It appears in the SaaS as “online” and sends heartbeats. Your firewall never opened.
Minute 9 — connect your data
A wizard collects host, port, credentials and paths. Those credentials stay inside the Runner — the control plane only keeps a logical reference. The Runner tests the connection and profiles the schema, so the agents see metadata, not rows.
Minute 15 — first scores
You pick a use-case type, import a template, and the data-science agent maps its expected features to your real schema and generates the code. The scenario runs in your data plane; results land in monitoring.
Why self-serve is the point
Self-service isn’t a cost-saving shortcut — it’s proof the architecture is sound. If getting live required our engineers on a call, the platform would be doing too little on its own. When you need a hand for tuning or validation, advisory is available by the day — but the starting line is yours to cross alone.
Start now on the onboarding page.