NDS AI

The right data for AI. Nothing more.

NDS keeps one live copy of your business data. Ask a question in plain English: AI writes the query, NDS runs it as you, and only the rows you may read come back.

AI assist Illustration with sample data
Ask as
You

Show me today's orders and who placed them.

NDS AI

Sure, let me do that for you. Here's the query I'm going to use to show you the result:

SELECT o.placed_at, c.name AS customer, c.region, o.amount
FROM sales.orders o
JOIN sales.customers c ON c.id = o.customer_id
WHERE o.placed_at >= CURRENT_DATE
ORDER BY o.placed_at DESC;

Ran as finance@example.com

17 rows from today

placed_atcustomerregionamount
11:42Nordlicht GmbHEU312.00
11:15Blue Harbor IncUS89.50
10:58Sakura WorksAPAC1,420.00
10:31Atelier RiveEU64.20

13 more rows

Reads customers, orders and payments in every region.

All five sources live together in the NDS Store, as one copy. This question needs orders and customers, and only the points each person may read move into the answer. The data itself stays in NDS.

Why does every new AI project add another system?

Each new use case tends to bring its own store, its own pipeline and its own rules. The data is the same; the copies multiply.

Another store for vectors

Search by meaning usually means a second database, with its own copy of your data to secure, back up and keep in step.

Another pipeline to keep in sync

Every copy needs a job that moves data, watches for changes and needs someone to look after it.

Another place for permissions to slip

Rules set in one system rarely follow the data into the next, so AI can end up seeing what people cannot.

AI does not need all your data. It needs the right part of it.

Hand AI too much and it gets slow, costly and more likely to guess. Hand it the wrong part and it still answers, with confidence. With NDS, a question narrows one copy of your data to the rows it needs, under the rules of the person asking, without copying anything first.

The question: “Show me today's orders and who placed them.”

An arrow made of data points. Its tail is the NDS Store, every source in one copy. Three gates narrow it: the sources the question needs, today's orders and their customers, and what the person asking may read. Points that are not needed fall away at each gate, and only a few reach the tip: 17 rows in the answer.
  1. NDS Store: everything, in one copy.
  2. Gate 1: only the sources the question needs.
  3. Gate 2: only today's orders and their customers.
  4. Gate 3: only what the person asking may read.
  5. The answer: 17 rows.
Illustration with sample data. AI asks once; NDS narrows the one copy to the rows the question needs, under the asker's rules. Nothing is copied along the way.

Search by meaning

Some questions are about meaning, not exact values. For those, NDS will find the items closest in meaning to the question, keep only the ones the person asking may read, and hand just those to AI. Your LLM writes the answer from them.

The question: “Which customers complained about late deliveries?”

Your data is a circle of points, one colour per source. Above it, a ring of small squares: each square is the vector for one item, and squares that sit close together mean similar things. The question becomes a vector too. It lands on the ring next to the items about late deliveries. One of those belongs to someone else, so it stays out. AI reads the three that remain, and the LLM writes the answer from them.
  1. Your question becomes a vector.
  2. The index finds the items closest in meaning.
  3. Only what you may read stays in.
  4. AI reads just those.
  5. The LLM writes the answer.
Illustration with sample data.

AI is only as good as the data it reads

AI assistants and agents now answer business questions and act on them. Most read a copy made for them: an export, a warehouse table, a document index. That copy is hours old and often lacks the access rules that protect the original. So the AI answers confidently, from the past.

TodayAI reads a warehouse copy

How many units of item B-117 can we still promise to customers?

You have 120 units available.

112 units were sold after 06:00. This answer is wrong.

Source: warehouse copy, as of 06:006 hours old
With NDSAI reads the live data

How many units of item B-117 can we still promise to customers?

You have 8 units available.

The answer matches what the business sees right now.

Source: NDS, as of 12:00Live
Illustration with sample data. The time now is 12:00.

Built into NDS

Capabilities that come with the database, not as extra systems to buy, connect and secure.

  • Bring your own model

    Use the AI model you choose: Ollama, Anthropic or any OpenAI-compatible endpoint, on your servers or in the cloud.

  • Ingestion pipelines

    Map, filter and preview data from MySQL, DuckDB and files, then run it now or on a schedule. It lands once and stays current.

  • Documents beside your data

    Store files and documents next to your rows, link them to records and search them by what they contain.

  • Plain English to SQL

    Ask a question in plain English. AI writes the query, shows it to you and runs it with your permissions.

  • ML training data tracking

    Keep model training data reproducible, so you can return to the exact data a model learned from.

  • History you can query

    See what changed in a table and when, and look at your data as it was before the change.

  • Row and column rules

    Decide once who may read which rows and columns. Apps, reports and AI questions all follow the same rules.

  • Your own programs

    Write your own logic, deploy it beside the data and run it on a schedule.

  • Says when it is out of date

    If the source has changed since the AI's material was prepared, the answer says so instead of staying silent.

All your clouds, in one place

Connect your clouds to NDS. Ask once — NDS reads them all as if they were one. AWS, Azure, GCP, your own data centre: connect them live to NDS. Ask in plain English. AI writes the query. NDS reads each connected cloud in place, joins the results with data it stores itself, and answers — nothing copied, every column labelled.

You · AI assist
NDS platform
NDSall your clouds · nothing copied
Query engine
ParsePlanOptimizeExecute
Pushes filters down to each cloud, so only matching rows travel.
Compute
Joins and totals run here, in parallel.
Storage
survey.responses survey.scores Row and column rules
ConnectorAWS · live
ConnectorAzure · live
Your clouds
Live
AWSFinance systemRead in place · nothing copied
revenuerenewalsinvoices
Live
AzureCRMRead in place · nothing copied
accountspipelineleads
A sales rep asking the same question sees only their own accounts.
Illustration with sample data. Numbers show how scores and results move together, not cause and effect.

One database, many tools and sources

Each name is a tool, technology or source NDS works with. Select one to see what you can do with it.

Select a name to see what you can do with it.

Business applications as sources

Bring data from the systems that run your business into the NDS Store, next to everything else, under the same rules.

ERP

  • SAP S/4HANA
  • Oracle Fusion Cloud ERP
  • Microsoft Dynamics 365
  • Oracle NetSuite
  • Infor
  • Epicor
  • Odoo

CRM

  • Salesforce
  • HubSpot
  • Microsoft Dynamics 365 Sales
  • Zoho CRM
  • ServiceNow

Finance

  • Workday Financials
  • SAP Concur
  • Sage Intacct
  • QuickBooks
  • Xero
  • Stripe

Streams and events

  • Apache Kafka
  • Change capture from databases
Ask for a source

Tell us which system you need first. Product names are trademarks of their owners; listing them does not mean a partnership.

AI agents that work within your rules

An agent should see what its person may see, and nothing more.

A business question can take an AI agent several steps. The plan is for every step to use the same current data, stay within the access of the person the agent works for, and leave a record you can review. Here, a sales assistant answers one question.

Sales assistant Live
Sales managerCan we promise Acme 500 units of A-204 by Friday?
Sales assistantYou can promise 420 by Friday and 80 by Tuesday. A supplier email says the next batch is 3 days late. Shall I reserve the 420?
Sales managerYes, reserve them.
Sales assistantStock just changed. I can reserve 408 now and 92 for Tuesday. Go ahead?
Sales managerGo ahead.
Sales assistantDone. 408 reserved for Friday and 92 scheduled for Tuesday. I'll tell you if stock changes again.
What the agent got
  1. Working for the sales manager, with the sales manager's accesspermitted
  2. Current stock for A-204: 612 on handcurrent
  3. Open orders for A-204: 192 already promisedcurrent
  4. A supplier email about the next batchrelated
  5. Supplier prices left out: the sales manager may not see themheld back
  6. Stock changed before the reservation, so the plan is out of dateplan expired
  7. Current stock for A-204: 600 on handcurrent
  8. Reserved 408 units for Acme, recorded for the sales manager and the assistantrecorded

Illustration. The figures are made up for the example.

Decisions NDS helps AI make

  • Promise dates customers can rely onStock, orders and supplier news in one current answer.
  • Stop fraud while a payment is still pendingLive payments beside the past cases they resemble.
  • Reorder before the shelf runs emptySales and stock checked as they change.
  • Send each case to the right personSimilar past cases, and who solved them.

What the agent gets

  • One consistent view of the data across every step
  • Only what its user is allowed to see
  • Earlier answers reused while the data is unchanged
  • Plans that expire when the data moves
  • A record of every step: the agent, and the person it acted for

Vectors and vector indexes, simply

Search by meaning finds what a question is about, even when the words are different. NDS will keep the vectors and their index inside the database, next to the data they describe, so a new record can be found without a copy step.

What is a vector?

A list of numbers that describes what something means. Texts that mean similar things get similar numbers, even when they use different words. Here, each square is one vector: similar shading, similar meaning.

What is a vector index?

A map of meaning. Instead of comparing a question with every item, it goes straight to the closest neighbours, the dashed circle here, and leaves the rest alone.

Live linkIllustration with sample data
Your dataEach record has its vector
    Meaning mapThe vector index
    Your LLMYour choice of model

    Ask a question above. The answer comes back here.

    Pick a question, or add a record and ask again.

    Illustration with sample data.

    A typical RAG setup, and the same with NDS

    Typical RAG setupRepeated on every change The LLM reads the last copy.
    1. Your data
    2. Export
    3. Re-embed all
    4. Vector database
    5. Rebuild index
    6. LLM
    With NDSData, vectors and index in one system The LLM reads the current data.
    Your dataVectorsVector indexone system
    LLM
    Illustration.

    Vectors live with the data

    No second database to copy into, secure or back up. The same permissions, backups and restores cover both.

    Only what changed is refreshed

    When a record or document changes, just its vectors are marked out of date and refreshed in the background. Nothing is copied again.

    Narrow first, then search by meaning

    NDS narrows by what it already knows, such as customer, region or date, then searches by meaning inside that small set.

    An index you can change freely

    The search index is kept apart from the vectors, so it can be rebuilt or swapped without recomputing a single vector.

    A clean switch to a new AI model

    Move to a new embedding model in one switch-over, never a half-old, half-new mix, and keep the old version for audit.

    Access checked at search time

    Search results follow the asking person's current permissions, not rules frozen when the index was built.

    One idea, every industry

    AI asks a question. NDS hands back a small result that is close in meaning, allowed for the person asking, and current. The idea is the same wherever data meets AI.

    • Payments
    • Genomics
    • Rail
    • Clinical trials
    • Satellites
    • Wealth
    • Luxury
    • E-commerce
    • Chemicals
    • Drones
    • Airlines
    • Grid
    • Renewables
    • 5G
    • Cloud
    • Customs
    • Public sector
    • Shipping
    • Warehousing
    • Streaming
    • Pharma
    • Capital markets
    • Education
    • Smart cities
    • Agriculture
    • Mining
    • Oil and gas
    • Real estate
    • Construction
    • Cybersecurity
    • Semiconductors
    • Gaming
    • Sports
    • Legal
    • Research
    • Hospitality
    • Travel
    • Data centres
    • Advertising
    • Water
    • Nuclear
    • Food and beverage
    • Hospitals
    • Marketplaces
    • Aviation

    Select an industry to see an example question.

    Examples show how NDS fits; they are not customer stories.

    Start small, grow into one copy

    NDS does not ask you to replace everything on day one.

    1. 1

      Start beside what you have

      Connect NDS to your current databases and ERP. Nothing is switched off.

    2. 2

      Point reports and AI at NDS

      Dashboards and AI assistants read live data from NDS, under your rules.

    3. 3

      Make NDS the one copy

      Move apps onto NDS when you are ready, and retire the copies you no longer need.

    Runs where you run

    Your laptop

    Try NDS locally. It ships as a container.

    Your data centre

    Run it on your own servers, fully in your control.

    Your cloud account

    Run it in the cloud you already use, next to your data.

    Security and governance by default

    The same rules apply whether a person, a report or an AI agent is asking.

    Encrypted connections

    Traffic between your tools and NDS is encrypted.

    Sign in with your identity provider

    Use the single sign-on you already have, and keep users in sync.

    Roles and privileges

    Decide who can read and who can write, in plain SQL.

    An audit trail

    Every request is recorded: who asked, what was read, and when.

    Knows who the AI worked for

    Each request records the agent that acted and the person it acted for, not just a shared service account.

    Answers you can reproduce

    Show what the AI was shown, when, and under which permissions and model, for as long as you choose to keep history.

    Data you can trust the moment it exists.

    A business should have one version of the truth, for people and for AI. NDS starts with the database, grows into the place where AI finds exactly what it needs, and goes all the way down to memory hardware built for data.

    See NDS in action

    Five ways to get in touch. Each one starts with a short form.

    Try NDS AI for free

    Want to see NDS work on your own data? Ask for a free trial and we'll help you get it running.

    • Free to try
    • Runs on your own computer or a Linux server
    • Load data from MySQL, DuckDB or CSV files
    • Ask questions in plain English with AI assist
    Request a free trial

    Request a demo

    See NDS answer questions on live data, and see the small result AI reads.

    • 30 minutes, online
    • Bring a question from your own business
    Request a demo

    Become a design partner

    Build NDS with us. We are looking for a small number of teams with real data and real AI questions.

    • Early access to NDS
    • A direct line to the founder
    • A say in what we build next
    Apply as a design partner

    For vendors

    Offer a product, data or a service that could help NDS or its customers? Tell us about it.

    • Tools, integrations and data
    • A short note is enough to start
    Contact us as a vendor

    For investors

    Every company is putting AI to work on its data. NDS is building the database that hands AI exactly what it needs.

    • Founder-led, early stage
    • Round and stage details
    Request the investor brief

    Anything else: support@ndsai.io

    Request a demo

    Tell us who you are and a question you'd like to see NDS answer.

    You're writing as

    Send opens your email app with this message ready to go to support@ndsai.io. Nothing is stored on this site.

    Log in

    For the NDS team.

    Forgotten your password? Email support@ndsai.io.