Your First Data Pipeline in 15 Minutes (Free Beta, No Warehouse Required)
Most teams stall on data not because they lack ambition, but because the first mile feels expensive: another SaaS bill, another consultant, another “platform” that assumes you already have a warehouse.
You don't need that to take the first step.
DataKru is in free beta with a simple path: one account, upload a file, run checks and a pipeline job, and see what ran. This post is that path—step by step, with the limits we enforce so there are no surprises.
If you want the same steps inside the product, open the 15-minute getting started guide after you register.
Who this is for
- Founders and operators with exports (CSV, spreadsheets) who want repeatable quality checks—not another slide deck.
- Solo data people who want ingestion, jobs, and lineage in one place without wiring five tools on week one.
- Curious practitioners evaluating whether an upload-first beta is enough for your current problem (often it is for learning and small batches).
Who it's not pretending to be yet: a multi-tenant enterprise platform with team RBAC, S3 connectors, or cron schedules. We say that upfront—see pricing & limits.
What you'll need
- 15 minutes and a browser.
- A CSV you are allowed to use (sample sales, marketing export, anonymized extract—nothing sensitive you shouldn't upload to a beta).
- No credit card. Fair-use caps apply (upload size, storage, concurrent jobs, jobs per day—listed on the pricing page).
The 15-minute path
1. Create an account (~2 min)
Go to datakru.com/auth/register.
One login per person today—shared team workspaces are on the roadmap. Use your own account; don't share passwords to simulate "teams."
2. Upload a file (~3 min)
Open Ingestion from the dashboard or tools/ingestion.
Upload your CSV. If you hit a limit, the UI should show a clear message (file too large, storage cap, etc.)—not a cryptic error. Those limits are documented in our terms and fair-use policy.
Beta honesty: connectors for warehouses and cloud storage are not in the product yet. Start with files; connectors come later.
3. Run a quality check (~3 min)
Open Quality (tools/quality) or follow the flow from your uploaded dataset.
The goal is simple: know if the file is usable before you chain transforms. Missing IDs, bad types, and duplicate keys are cheaper to fix here than in a downstream dashboard.
4. Configure and run a pipeline (~5 min)
Open Pipeline (tools/pipeline).
Pick your uploaded file, add stages (quality, standardization, transform—what your data needs), and run manually. Automated schedules (hourly/daily cron) are not live in beta; the UI reflects that. You trigger runs when you want them.
This is deliberate: we'd rather you trust manual runs today than think a schedule exists when it doesn't.
5. Confirm in Jobs (~2 min)
Open Jobs (tools/jobs).
You should see your run—status, timing, and enough detail to answer "did it finish?" in a standup. If something fails, you get a failure state to debug, not silence.
Optional stretch goals if you have time: peek at Lineage (tools/lineage) or ML Studio (tools/ml) with the same account. Not required for the first 15 minutes.
What "good" looks like after 15 minutes
You can honestly say:
- I have a file in the workspace.
- I ran a quality pass.
- I executed a pipeline job and verified it in Jobs.
- I know where fair-use limits are documented.
That's a first pipeline, not a production platform. For many teams, that's the right wedge: prove the workflow before you buy infra.
Honest limits (read this once)
We'd rather lose a signup than mislead you.
| Topic | Beta today |
|---|---|
| Sources | File upload primary; no S3/Postgres connector in-product yet |
| Teams | One account per person; no RBAC |
| Schedules | Manual runs; cron coming later |
| Jobs | Run in the app stack; long-running work is improving (dedicated worker on the roadmap) |
| Support | Best-effort email (hello@datakru.com); not 24/7 SLA |
| Assistant | Rules-based commands on the dashboard; voice/ASR may be unavailable in your browser |
Full detail lives on the homepage beta note and pricing.
When to go further
Stay on free beta if you're learning, prototyping, or running occasional batch files.
Have a complex project in mind? Multi-source ingestion, production architecture, tenancy, connectors, or an ML/data platform you can't fake in a 15-minute walkthrough—book a Data & AI strategy call and we'll scope it honestly (what fits the product today vs a structured engagement):
You can also use the architecture sprint inquiry on our pricing page for written follow-up.
Try it now
- Create a free account
- Follow the 15-minute guide
- Stuck on the beta path? Support or hello@datakru.com
- Bigger build? Book a strategy call
If you complete the 15-minute path, we'd love to know what broke—or what clicked. Reply on X @datakruHQ or email hello@datakru.com. We're building in public; your feedback shapes the roadmap.