LumaVista is a research platform that turns a question into a structured, multi-source report — without you having to open dozens of tabs or piece together findings manually. You describe what you want to know, and a team of specialized AI agents plans the research, searches across sources, reasons through what they find, and writes a synthesized report.

This guide covers creating your account, understanding the dashboard, and getting oriented before your first project.

Creating your account

Getting started takes about a minute:

Visit lumavista.eu and click Sign Up.
Enter your email address and choose a password. That is the whole form. If your organisation uses single sign-on, enter your work email and LumaVista will hand you off to your identity provider instead — see SSO for organisations below.
Set up your encryption. This is the step that matters, and it happens right after the form. LumaVista generates an encryption key on your device and shows you a recovery sheet — a printable page with a QR code. Save it. Details in How Your Data Is Protected.
Confirm your email. Check your inbox for a verification link.

Once you are in, you land on the dashboard — the home base for all your research.

Signing in later

Three ways in, all from the same login page:

Email and password — the usual route.
A link by email — enter your address and LumaVista mails you a one-time sign-in link. This is also the answer if you have forgotten your password: sign in with the link, then set a new password in Settings → Account.
SSO for organisations — if your organisation has configured single sign-on, entering your work address routes you to your identity provider. This is an organisation-level setup, not something you switch on yourself.

There is no “sign in with Google” or “sign in with GitHub”. You can connect Google, Slack or GitHub to LumaVista later so that workflows can act in those tools — but that is authorising LumaVista to reach them, not using them to log in.

The dashboard at a glance

You can look at the real thing before reading about it. The live demo drops you into a read-only session on an actual account with finished research in it — a real project tree, a real knowledge graph, a real report. No signup, and nothing described from memory.

The dashboard is designed around one idea: every research effort is a project. When you log in, you see your project list front and center — cards showing each project’s title, status, and when it last ran.

Here is what you will find:

Project cards — Each card represents a research project. Cards show the project title (derived from your prompt), its current status (running, completed, or stopped), and a brief summary. Click any card to open it.
New Project button — Top of the project list. This is where every research session starts.
Sidebar — Navigation to your projects, settings, and usage information.
Settings — Account, appearance, research defaults, privacy and encryption — 19 tabs with a search box.

If this is your first time logging in, your project list will be empty. That is expected — we will fix that shortly.

What is a project?

A project in LumaVista is a self-contained research effort. You give it a goal — a question, a topic, a problem statement — and LumaVista assigns a team of AI agents to investigate it. Each project has:

A prompt — The research question or goal you provide. This is the starting point for everything the agents do.
A research graph — A visual representation of the work being done. Agents create nodes in this graph as they plan subtasks, run searches, reason through findings, and synthesize results. You can watch this graph grow in real time.
A report — The final output. Once agents finish their work, you get a structured report that synthesizes what they found across all their searches and reasoning steps.
A chat panel — A way to interact with your project after (or during) execution. Ask follow-up questions, request changes, or dig deeper into specific findings.

Projects are persistent. You can come back to a completed project days later, read the report again, or ask new follow-up questions in the chat panel.

Your first steps

Here is the recommended path for new users:

Create your first project. Click “New Project” on the dashboard. You can type a research question directly, or use the guided prompt builder to refine your question through a short conversation. We recommend starting with the guided prompt builder — it helps you articulate exactly what you want researched.
Watch the agents work. Once you launch a project, the research graph appears. You will see planning nodes break down your question into subtasks, search nodes gather information, and reasoning nodes analyze what was found. This usually takes a few minutes depending on the complexity of your question.
Read your report. When the agents finish, your report is ready. It is structured with sections, citations, and synthesized findings — not just a dump of search results.
Ask follow-up questions. Open the chat panel and ask about anything in the report. Want more detail on a specific section? Curious about something the agents did not cover? The chat understands the full context of your research.

Choosing a research strategy

LumaVista does not have one research engine with a speed dial. It has a catalog of strategies — different research shapes for different questions — and it picks one for you based on what you asked.

When you describe what you want, LumaVista analyses the question and proposes a strategy, shown as a set of ranked options you can accept or override. The published strategies are:

Strategy	Shape
Landscape Survey	Broad scan of a field. Who is in it, what are the themes.
Scout & Dive	Scout the territory first, then investigate the promising parts in depth.
Deep Research	Maximum coverage — parallel investigation of every dimension, then iterative gap-filling until coverage holds up.
Recursive Deep Research	Deep Research plus signal-driven loops: it chases leads it did not plan for and resolves contradictions it finds.
Iterative Deepening	Start shallow, go deeper in rounds, stop when the returns flatten.
Comparative Analysis	Structured comparison of named options against shared criteria.
Fact Check	Verify specific claims against sources, with adequacy checks on the sources themselves.
Corporate Due Diligence	Structured investigation of a company across financial, legal and operational dimensions.
Person Background	Structured background research on an individual.
Cyber Footprint	Externally visible technical and organisational surface.
Investigative Journalism	Lead-driven investigation with an emphasis on corroboration and provenance.

Most strategies then ask you one or two questions of their own before starting — thoroughness, and what shape the output should take. Landscape Survey, for instance, asks whether you want a quick scan or a standard survey, and whether the result should be a written report, a bullet summary, or a table.

You are not locked in. The strategy is chosen per project, not set once as an account-wide default.

A note on privacy

Your research data is encrypted and stored per-user. LumaVista uses a per-user storage architecture — your data lives in its own isolated database, not mixed in with other users’ data. The encryption key is generated on your device at signup and LumaVista never holds it in the clear — see How Your Data Is Protected.

Next steps

You are ready to run your first project. Head to Your First Research Project for a step-by-step walkthrough of creating a project from prompt to finished report.

If you want a deeper tour of the interface first, check out Navigating the Interface for an annotated walkthrough of every panel and control.

Research runs are metered in credits. How Credits Work covers what spends them, where to watch your balance, and how LumaVista tells you a cost before you commit to it.

As you build up research over time, LumaVista automatically captures what you learn into a personal Knowledge Graph that makes future projects sharper and more relevant.